00:00:00
Welcome, welcome everyone to Wharton
00:00:02
Moneyball, the show where sports,
00:00:04
statistics, and business all intersect.
00:00:06
Three of my favorite topics. This is
00:00:08
Eric Bradlo, professor of marketing and
00:00:10
statistics and data science here at the
00:00:11
Wharton School. Some combination of
00:00:14
myself, my co-host and friend today,
00:00:16
Audi Winer, are here every week.
00:00:18
Sometimes it's us two, Cade Massie,
00:00:21
Shane Jensen, but this week it's Eric
00:00:23
and Audi here on the Wharton Podcast
00:00:25
Network. I think for those of you that
00:00:28
have been on our show or been with our
00:00:30
show for the last 11 plus years know
00:00:32
whenever it's me and Audi, I take this
00:00:34
opportunity to in some sense interview
00:00:37
Audi. So today there will not be a
00:00:39
guest. I will interview Audi for if
00:00:41
you'd like an hour or so during our
00:00:43
podcast. The first part of the show will
00:00:45
be our standard what caught your eye in
00:00:47
sports segment. Then the second half
00:00:49
you're all in for a real treat. As I
00:00:51
think many of you know, Audi, besides
00:00:53
being the co-f faculty director of
00:00:55
Wasabi, this Wharton Sports Analytics
00:00:57
and Business Initiative, which is part
00:00:58
of the bigger umbrella brand that I run,
00:01:00
the Wharton AI and Analytics Initiative,
00:01:03
also it runs our faculty research with
00:01:06
our undergraduates, and he's going to
00:01:08
talk to us about the research they're
00:01:09
doing in sports analytics. So, the first
00:01:11
half will be what caught your eye in
00:01:12
sports. The second part will be what
00:01:15
Audi is doing with our brilliant Penn
00:01:17
undergraduates on sports analytics and
00:01:19
research. Audi, how you doing today?
00:01:22
>> I'm doing really well. Excited to have a
00:01:23
good conversation with you, Eric. I know
00:01:25
that it's going to be a tremendous
00:01:27
temptation for us to spend the entire
00:01:29
time talking about the Yankees, but we
00:01:30
won't do that um at least not
00:01:32
exclusively. And I look forward to
00:01:34
talking to you about the work that our
00:01:36
students are doing, which are not not
00:01:37
only includes uh undergrads, but also um
00:01:39
graduate students, our PhD students, as
00:01:41
well as some students who are masters in
00:01:43
data science, and we even have a PhD or
00:01:46
two from other departments. Well, I know
00:01:48
at one point I think Ryan Bro was AMCS
00:01:50
or some students were AMCS.
00:01:52
>> Ryan was AMCS. Um, he was in applied
00:01:55
math and now he's with the Utah Jazz. He
00:01:58
actually came and visited. He he talked
00:01:59
about one of his um uh he talked about
00:02:02
some of the the uh the difficult
00:02:03
problems in in general.
00:02:05
>> Well, we'll get to that. We'll get to
00:02:06
that in the uh second half of our show.
00:02:08
So, what I wanted to start with today
00:02:10
was the NBA.
00:02:13
So, something very interesting is
00:02:16
happening in the NBA.
00:02:18
So, and it's maybe the most
00:02:20
extraordinary start to a season that
00:02:22
I've ever seen. I would just like your
00:02:24
take on it from a statistical
00:02:25
perspective. A very specific question.
00:02:28
So, the defending champions, the
00:02:29
Oklahoma City Thunder,
00:02:32
they won the title last year, as you may
00:02:33
remember.
00:02:34
>> I do remember.
00:02:35
>> They are 17-1
00:02:38
to start the season.
00:02:41
Right now, the forecast number of wins
00:02:43
for them is 68.
00:02:46
Now, you might say, "Well, that's a huge
00:02:48
number, maybe, except they're 17-1." If
00:02:52
you just projected that out, I'm not
00:02:54
saying a linear projection. Multiply it
00:02:58
by three, they would be 51 and three in
00:03:01
their next 54 games. Now, we all don't
00:03:04
predict that, but for them to beat that
00:03:07
prediction of 68, they just need to go
00:03:10
better than 51 and 13. So, let's be
00:03:13
clear. They're on a 51 and3 pace right
00:03:16
now.
00:03:19
51-13 gets them to 68 wins. So, why
00:03:22
don't we take it piece by piece?
00:03:25
Would you go over 68 knowing, by the
00:03:29
way, you may remember the highest number
00:03:31
of all time is 73. That was the Golden
00:03:34
State Warriors of I think 2017 who lost
00:03:36
in the finals, you remember, to LeBron.
00:03:38
>> They were 73 and N. Of course, the
00:03:40
Michael Jordan Bulls, I forget which of
00:03:42
the years was 72-10.
00:03:45
I think that might be maybe there's one
00:03:48
other at 70, but the next gap I know
00:03:50
there's a 69. So, let's just start with
00:03:53
that. I've got two other things to say
00:03:54
about the Thunder right now. How would
00:03:57
you help our listeners here on Wharton
00:03:59
Moneyball on the Wharton podcast network
00:04:01
think about how likely is it above 68
00:04:04
and 14 or possibly even the record of
00:04:06
all time?
00:04:07
>> Well, you know, there's two broad ways
00:04:09
to approach this. One is the way you did
00:04:12
already, which is just look at what are
00:04:14
the best seasons ever and essentially
00:04:17
ignore the what we any individual
00:04:20
information we have about OKC. They've
00:04:22
won 17 and lost one. Okay, we'll just
00:04:24
put that in the bank and then we'll use
00:04:26
that um and we'll ignore that and we'll
00:04:28
just say how likely is it that they're
00:04:29
going to be one of these top five teams
00:04:31
ever? And I guess 68 wins would put them
00:04:33
what seventh, eighth best team ever.
00:04:35
>> Yeah, probably exactly in that range
00:04:37
>> and considering that uh and that and
00:04:40
that essentially asks that question in
00:04:42
that way and which case we are really
00:04:44
not really thinking too much about the
00:04:46
individual team. were just saying they
00:04:47
went 17-1. All the the all the all the
00:04:50
data suggests they're they're
00:04:52
potentially all-time great team. Where
00:04:53
would you put them? 68 seems reasonable.
00:04:56
Um I I don't think that's where it came
00:04:58
from. My guess is the estimate came from
00:05:01
some sort of basian updating, which is a
00:05:03
a tool that we've done a lot in this in
00:05:05
this program. You essentially shrink or
00:05:08
regress to the mean um to estimate what
00:05:11
you might call their true win rate. So
00:05:13
there we don't believe they're really a
00:05:15
17 out of 18 team which is like a 93%
00:05:18
win rate. We think they're something
00:05:20
lower than that. And the question is by
00:05:22
how much do we regress down to the mean.
00:05:24
And so we're essentially predicting that
00:05:26
they go what uhif what you say 51 and
00:05:29
>> 51 and 13.
00:05:31
>> So that's their like 75%
00:05:35
>> little 80%. So we're saying the rest of
00:05:37
the way. So they're 93% win up until
00:05:40
now. The best estimate of what their
00:05:42
true talent is is 80% and that leads to
00:05:45
68. That's probably what they did,
00:05:47
right? So would you would you be would
00:05:49
you reasonably treat this as you know
00:05:51
the classic beta binomial situation and
00:05:54
what I mean by that is we have a prior
00:05:57
for OKC. We have a prior for every team.
00:06:00
Now obviously this this part I can do
00:06:03
this it's not even math it's intuition.
00:06:05
If they're 17 and one, which is 93%.
00:06:08
Okay. And the prediction is that they're
00:06:10
going to go 80% the rest of the way.
00:06:14
>> The prior must be below 80%. Because if
00:06:19
the prior were above 80% and the
00:06:21
likelihood were above 80%, which we know
00:06:23
it's 93%, the posterior would have to be
00:06:27
a convex combination of those two, which
00:06:30
means that the prior for the Oklahoma
00:06:32
City Thunder might have been, let's even
00:06:33
say it was 75%. Which is not
00:06:36
unreasonable. Maybe they were projected
00:06:37
to be a 6061 win team, which is 75%.
00:06:42
They're 17-1. We're now up to a
00:06:44
prediction as you pointed out of about
00:06:46
80% and that's our prediction for the
00:06:48
rest. That sounds about right, doesn't
00:06:49
it?
00:06:50
>> Yeah, it does. But I don't think so.
00:06:52
There's two ways to do this this basian
00:06:54
this shrinkage. Do you shrink to a prior
00:06:56
that's specific to the team or do you
00:06:59
shrink to the league overall prior and
00:07:00
the difference is a prior shrunk to the
00:07:03
specific team has very has much smaller
00:07:05
variance. Right? So if we are if we're
00:07:08
talking about OKC given what we know
00:07:10
about everything what they did up until
00:07:11
this season, we probably have a 75%
00:07:14
centrid maybe uh or posterior or prior
00:07:17
mean and probably a pretty small
00:07:20
standard deviation. Um and therefore um
00:07:24
we ignore we shrink pretty heavily
00:07:25
despite the fact they got 17 and one. We
00:07:27
still shrink pretty heavily back to the
00:07:29
prior. Just put it up a little higher.
00:07:31
The other way that you could do it is
00:07:33
just ignore, pretend this is the only
00:07:35
thing you know about this team is that
00:07:37
it went 17 to1. That's it. You know
00:07:39
nothing about the previous years. And
00:07:41
then you shrink to 50%. But your but
00:07:43
your prior mean would be huge, much
00:07:45
bigger. And so the current data would
00:07:48
end up getting much more weight. Um so
00:07:50
you have two choices. You either shrink
00:07:52
heavily to a very high mean or you
00:07:54
shrink.
00:07:56
>> I'm shrinking towards the OKC one, but I
00:07:59
do agree with you. Either one of those
00:08:01
conceptually depending on how much
00:08:03
sample size you put in the prior could
00:08:06
lead to the 80% number in this case.
00:08:08
>> Either one. It's funny how it's funny.
00:08:10
It's interesting how to do this. It
00:08:11
>> is interesting.
00:08:12
>> So, um I probably would would have had a
00:08:15
a multivariant. I would have lumped
00:08:18
other teams into into that good team
00:08:20
prior instead of just prior and try to
00:08:23
borrow something else. Um but that's not
00:08:25
I don't want to try to do this myself.
00:08:27
We're trying to predict them whether or
00:08:28
not. So essentially, I guess the prior
00:08:30
one would get back get to an 80% going
00:08:32
out. Um I think that um 68 I think
00:08:37
they're going to go over. That's my
00:08:38
that's my point. I
00:08:39
>> think so too. So there's something else
00:08:41
about them that's interesting.
00:08:43
>> Right now their projection
00:08:47
is to win the league by 10 wins. That
00:08:52
means they'll have 10 more wins than any
00:08:54
other team. Now, that would be a
00:08:56
historic
00:08:57
>> historic mark. Well, can I ask you one
00:08:59
question? A particular question. I don't
00:09:00
know. Has their schedule been average,
00:09:03
difficult, easy so far? What do we know?
00:09:05
>> It's a good question. Um, somebody
00:09:07
knows.
00:09:08
>> Yeah.
00:09:08
>> This somebody doesn't know, but the
00:09:11
answer is I don't know. That's a fair
00:09:13
question. I mean, how I mean, we're
00:09:15
starting to get to enough that you and I
00:09:18
would agree. It's hard to believe it's
00:09:20
much below 500, if it's at all below 500
00:09:23
of the teams they played, right?
00:09:25
Because, you know, eventually it's
00:09:28
eventually going to average out to a
00:09:30
certain schedule.
00:09:31
>> Well, there's the problem is in the NBA
00:09:33
there aren't that many great teams.
00:09:34
>> No, no, that's the problem, right? So,
00:09:36
your point is there's no information.
00:09:38
Like, I hate to make this up, but they
00:09:41
could be equivalently like five and one.
00:09:44
Like, of course they beat those 12 teams
00:09:46
and then there's like six games. All
00:09:48
right. You went five and one in in tough
00:09:50
games. All right. Well, all right.
00:09:51
That's
00:09:52
>> Now, they played How many games going
00:09:54
ahead do they have against really good
00:09:56
teams?
00:09:57
>> I haven't I haven't looked at that. And
00:09:58
by the way,
00:10:00
>> another way to do the prediction, which
00:10:02
is we actually build a simulator game by
00:10:05
game. We play that out and then it takes
00:10:09
into account who they're playing, their
00:10:10
relative strength.
00:10:12
>> That's that's the way to do it. The the
00:10:14
difficulty in doing that simulation is
00:10:16
that you have to put in some sort of
00:10:18
talent metrics for for everyone. And uh
00:10:21
we've been talking about this on our
00:10:22
show with Cade on how to simulate going
00:10:25
forward. Do you either put the
00:10:27
uncertainty in the parameter estimates
00:10:30
before you start the sim or do you treat
00:10:32
the the the parameters as fixed and then
00:10:36
you just reestimate them as you go
00:10:37
through the sim.
00:10:40
If you are fully basian that those
00:10:41
produce the same and are correct about
00:10:43
the priors that produce the same
00:10:45
answers. Um if you are not fully basian
00:10:48
if you just and you just want to add
00:10:50
uncertainties to some say MLE then uh
00:10:53
then you'll get very different results.
00:10:55
>> Agree with that. Let me point out
00:10:56
something though that um people might do
00:10:58
though for practical reasons and then
00:11:00
we'll move on. This is more of a
00:11:01
technical comment here, but the first
00:11:03
one you mentioned where you fix the
00:11:05
parameters but then add the simulation
00:11:07
error later as you go through because
00:11:09
you up you rerun it kind of that one of
00:11:12
course will um take potentially less
00:11:16
computation because you've batch
00:11:18
processed the data and now all and
00:11:21
you've generated a posterior and now
00:11:23
you're just adding one observation at a
00:11:25
time which could there other ways all
00:11:27
I'm pointing out is there could be
00:11:29
computational iational advantage to do
00:11:31
sequential simulation but it also
00:11:35
depends on what your algorithm is. So if
00:11:36
you're if you have just some algorithm
00:11:38
that estimates team strengths
00:11:40
>> oh then that's fine and that can be done
00:11:41
fast and then you're then you have no
00:11:43
then there's no advantage. Just one
00:11:44
other thing about the Thunder before we
00:11:46
move off the NBA.
00:11:48
>> We talked about this last season in some
00:11:50
metrics they had the greatest season in
00:11:52
history last season. Forget number of
00:11:53
wins. Their point differential broke the
00:11:55
record. I think it was somewhere around
00:11:57
11.5.
00:11:59
So right now, Audi, they're at 16.9.
00:12:02
>> Ridiculous.
00:12:04
>> So
00:12:06
at some point, we're going to have like,
00:12:09
let's say they win the title again this
00:12:10
year. Let's say they win 70 plus games.
00:12:14
Let's say they have a point differential
00:12:15
of 13 plus. We're going to have to start
00:12:19
talking about this OKC team as being
00:12:22
it's not a dynasty yet. Two doesn't
00:12:24
necessarily get you to some definition.
00:12:27
We're going to have to start talking
00:12:28
about them as one of the greater teams.
00:12:31
I don't know the top 5% of teams all
00:12:33
time. I mean, if this happens,
00:12:36
is it? And but the trick is to me, yes,
00:12:38
of course we would. But my question is
00:12:40
is a basketball one. What is the
00:12:42
anomaly? is the anom because they don't
00:12:45
have is it the anomaly that uh she is
00:12:48
really a great player like he's a LeBron
00:12:50
level do you know alltime Michael Jordan
00:12:53
level and we've just underestimated him
00:12:54
or is it a team that's extraordinarily
00:12:57
well constructed that has no weaknesses
00:12:59
that have that have found that defense
00:13:01
is extremely important and while it does
00:13:03
it's not flashy it's the way you make
00:13:05
big points
00:13:07
>> if you're asking me as a basketball fan
00:13:09
what I would tell you is they're an
00:13:11
extraordinarily well constructed ed team
00:13:14
that Shay Gildish Alexander is the best
00:13:16
player or top two or three in the NBA
00:13:18
right now, but he's not Michael Jordan
00:13:21
level talent or
00:13:22
>> prime for or even Seth Curry at his
00:13:24
prime or
00:13:24
>> right probably not. But he might be in
00:13:26
that next tier down and then they've got
00:13:29
a bunch of very very good other players
00:13:32
and so you take a great player and a
00:13:35
bunch of very very good other players
00:13:37
and you might get there and they seem
00:13:39
extremely well balanced. So either way,
00:13:41
I don't want to spend all of our time
00:13:43
talking about the NBA. I was just
00:13:44
noticing their records incredible. I
00:13:47
glad we talked about some different ways
00:13:48
to do prediction, whether it's through
00:13:50
shrinking to their prior, shrinking to
00:13:52
the league prior, a simulationbased
00:13:54
prior, but also this point differential
00:13:58
is just incredible right now. And at
00:14:00
some point we have to say, I understand
00:14:01
it's only 18 games. Okay. So, next week,
00:14:03
Audi, if we're sitting here and it's 23
00:14:05
games and they've got a 16 17 point
00:14:08
differential, eventually we're going to
00:14:10
start to have to say we'd be surprised
00:14:12
if they didn't break the overall record.
00:14:14
>> Yeah. I mean, not to say I I don't have
00:14:16
the the number exactly off hand, but I
00:14:18
believe the RMSSE on predicting wins
00:14:21
given your point differential is three.
00:14:24
Meaning that I can predict your total
00:14:26
wins just using your point differential
00:14:28
to win at about plus or minus three
00:14:29
wins. But how well do you think that'll
00:14:31
work in the tails?
00:14:33
>> How do you think?
00:14:34
>> So the the data that I fit it to, right?
00:14:37
So which is every season up until this
00:14:39
year doesn't seem to have a nonlinearity
00:14:42
at the tails. It's it's not drifting.
00:14:45
>> Um
00:14:45
>> there has to be some because there is an
00:14:47
upper bound. You can only win so many
00:14:49
games,
00:14:50
>> right?
00:14:51
>> It's predicting. It just seems to I mean
00:14:53
it's not it's I'm not predicting. So I
00:14:56
just predict wins. I mean, you could
00:14:57
predict winning percentage and then do a
00:14:59
logistic uh and then predict a logistic.
00:15:01
That's those are two I now we're getting
00:15:03
really in the weeds here for our show. I
00:15:04
don't want to I don't want to get into
00:15:06
this. Um but just to say that if they
00:15:08
are if they can hold 16, they are going
00:15:11
to break that record easy.
00:15:12
>> I think it's easily. Yeah. All right.
00:15:14
So, let's move on. Let's talk a little
00:15:16
about the NFL. Now, of course, we could
00:15:18
talk about the Eagles game and all that,
00:15:19
but I don't want to talk about that for
00:15:20
just a second. So, I just downloaded the
00:15:23
uh ESPN
00:15:26
power rankings for the NFL. Okay.
00:15:30
What team do you think is number one?
00:15:32
>> Rams.
00:15:34
>> That is incorrect, sir.
00:15:35
>> Is it still the Eagles?
00:15:37
>> Nope.
00:15:40
>> Colts.
00:15:42
So, so far you said the Rams, they're
00:15:43
second. You said the Eagles, they are
00:15:46
sixth. The Colts are five.
00:15:51
Who am I missing here?
00:15:54
The the Patriots. [laughter]
00:15:57
>> They're not even They're like 15 or 16.
00:15:59
>> Yeah. I mean,
00:16:00
>> it's the Chiefs.
00:16:01
>> The Chiefs still got them up there.
00:16:03
>> I I This is the part I This You have to
00:16:06
expl So, I know this I know this is
00:16:09
current. Let me just be clear. I know
00:16:10
this is current because it has their
00:16:12
record at six and five. That's their
00:16:14
actual record. Now,
00:16:16
>> it has them at 7.1 points above average.
00:16:20
It has the Rams at six. So that means
00:16:23
they're a one-point favorite against the
00:16:24
Rams. The Packers are third, but let's
00:16:26
say they played the Eagles. Right now,
00:16:28
it has them as a 2.7. Let's say they
00:16:30
played the Patriots. It has them as a
00:16:33
6.6 point favorite on a neutral field.
00:16:39
Well, um what what are the underlying
00:16:41
stats? I heard one stat that I'll share
00:16:43
with you.
00:16:44
>> Okay.
00:16:44
>> Um that uh I was trying to make some
00:16:46
heads or tails of this. I saw this on uh
00:16:48
sports analytics Twitter. Um um someone
00:16:51
pointed out that the the KC, the Chiefs
00:16:56
have the highest uh um number of yards
00:16:59
per drive.
00:17:02
It's a so basically how many yards are
00:17:04
they averaging per drive? Um and they're
00:17:07
just about 40. And and they showed then
00:17:09
the teams at that level historically go
00:17:11
12 and you know 12 and four or 13 and
00:17:14
three. Just so you know, by the way, I
00:17:15
think you remember last year that Chiefs
00:17:16
were something like 11-0 in one score
00:17:19
games. Just so you know, they're at
00:17:21
least I I well, I guess it just changed
00:17:23
up until this last week where they beat
00:17:25
the Colts, they were 0 and5. They're now
00:17:28
one and five in one score games.
00:17:30
>> One score games. So then so this this
00:17:33
particular stat suggests that this
00:17:35
underlying metric which doesn't take
00:17:37
into account special teams and and uh
00:17:40
things like that and and position
00:17:42
starting positions. So, for example,
00:17:44
you'll have more you'll expect more
00:17:45
yards on your drive if you start deep in
00:17:47
your in your own territory. You have
00:17:48
more more room to go. Um, and uh I don't
00:17:53
know whether that number means anything.
00:17:54
It's like a peripheral that's correlates
00:17:56
with with team quality, but they were
00:17:58
number one in it. And and usually teams
00:18:00
that historically teams with that level
00:18:02
of success in terms of yardage have been
00:18:05
have been great teams. So they they're
00:18:07
essentially that one number suggests
00:18:09
that the underlying metric suggests that
00:18:10
the KC is much much better than they
00:18:14
than they appear based on their record.
00:18:15
They've lost a bunch of one-run games.
00:18:17
They've had some bad special team
00:18:19
turnovers. They've had some what you
00:18:21
might just jump into is and lump into
00:18:23
one big pile called bad luck. And bad
00:18:26
luck is not supposed to continue. You
00:18:28
should expect neutral luck. So, if their
00:18:30
underlying posh strength parameters are
00:18:33
excellent, you might and given their
00:18:34
historical performance and and of course
00:18:36
who their quarterback is and the fact
00:18:38
that everybody seems to be beating up on
00:18:40
everybody else, maybe they do land that
00:18:42
land that high. But I have to tell you,
00:18:44
I'm surprised. I am surprised to by the
00:18:46
way, the only other team that's six and
00:18:48
five in the top like 12 teams, not
00:18:52
surprisingly, and this might be the role
00:18:54
of Prior. Well, who would you guess
00:18:55
who's the other six and five team that
00:18:57
the Priars are going to bring Ray up to?
00:19:00
Who?
00:19:00
>> Bills.
00:19:02
>> That's a great question. It's not the
00:19:03
Bills. The Bills are seven and four, by
00:19:05
the way.
00:19:06
>> Four. Okay.
00:19:06
>> It's the Ravens.
00:19:07
>> The Ravens, right? There's the Ravens.
00:19:09
Another one. They they they also have
00:19:11
prior history. And I wonder what this is
00:19:13
ESPN. The ESPN might have a lot of
00:19:15
weight on
00:19:16
>> Well, that's why I was asking you
00:19:17
because I think most people would find
00:19:19
it shocking that the Chiefs are the top.
00:19:22
I think mo a lot of people might find it
00:19:23
surprising that essentially the Eagles
00:19:25
and Ravens are equal. The Ravens are
00:19:28
better than the 8 and3 Seahawks, the
00:19:30
seven and four ners, the 7-4 Bills. The
00:19:33
9 and2 Broncos are like 12th.
00:19:38
>> The Patriots at 10 and two are like
00:19:41
15th.
00:19:42
>> Yeah. Yeah.
00:19:43
>> So either way, I thought it was just
00:19:45
interesting to talk about that.
00:19:46
>> They made a dip twice probably.
00:19:48
>> By the way, let me just say for all of
00:19:50
our sports, so as everyone knows, we
00:19:51
record Wharton Moneyball on Tuesday.
00:19:53
This is Tuesday, two days before
00:19:54
Thanksgiving.
00:19:56
There's an amazing schedule on
00:19:58
Thanksgiving, Audi. I don't know how
00:20:00
much you're going to get to watch. Um,
00:20:02
the first game is Lions and Packers. I
00:20:04
mean,
00:20:05
>> that's a great game.
00:20:06
>> That's a big game.
00:20:07
>> Lions traditionally play on
00:20:08
Thanksgiving. I guess there's more games
00:20:10
than one now. They used
00:20:10
>> the Lions always do, but there's they're
00:20:13
seven and four and the Packers are 73
00:20:14
and one. The second game all of a sudden
00:20:17
now became fascinating. Audi Chiefs at
00:20:20
Cowboys.
00:20:21
>> Wow. With the Cowboys beating the Eagles
00:20:23
last year.
00:20:23
>> So, six and five. I know with six and
00:20:27
five Chiefs against five five and one
00:20:29
Cowboys. A week ago was like they're
00:20:31
both going to lose. It's going to be two
00:20:32
losing teams. Now all of a sudden these
00:20:34
teams are in it. And the night game is
00:20:37
an interesting one because Joe Burrow's
00:20:38
coming back. It's Bengals at Ravens. So
00:20:41
all of a sudden we actually have three
00:20:43
sort of interesting games on
00:20:44
Thanksgiving Day. Hey, one thing I'm
00:20:47
going to take advantage of because it's
00:20:48
you and me here. So I I want to talk to
00:20:50
you about the Hall of Fame, the baseball
00:20:51
hall of
00:20:52
>> course. How could WE NOT, ERIC? IT'S OUR
00:20:54
DREAM.
00:20:54
>> We're go. We're going to But okay, so I
00:20:58
want to remind everybody there's two
00:21:01
separate committees that are going to be
00:21:03
voting this year. Okay.
00:21:06
One is what's called the contemporary
00:21:09
era
00:21:11
committee which used to be called I
00:21:14
don't know if it was the oldtimers
00:21:15
committee or anything but there's
00:21:16
>> veterans committee. Was it just
00:21:17
>> veterans committ? No. No. But they've
00:21:18
split it up. Remember Audi? There's 1980
00:21:21
onwards which is the contemporary era
00:21:24
candidates and then there's pre980.
00:21:27
So here are the seven candidates in the
00:21:30
contemporary era committee. Okay.
00:21:35
Roger Clemens.
00:21:38
So we agree he's and by the way you need
00:21:40
75% I think of like 16 voters. So not
00:21:46
>> he's not going to get it. Okay. Carlos
00:21:48
Delgado.
00:21:50
>> Carlos Delgado. I don't I don't see him
00:21:53
getting it. I don't I wouldn't imagine
00:21:54
he deserves it either. How about you?
00:21:55
What do you think?
00:21:56
>> 473 home runs.
00:21:58
>> Yeah.
00:21:59
>> Uh three time silver slugger. I don't
00:22:02
know. Right on the border. Jeff Kent.
00:22:06
>> On the border also. You know, this is a
00:22:08
Yeah. I mean, Kent, uh Kent, what was
00:22:11
he? His was an middle infield position,
00:22:13
wasn't he? Kent.
00:22:14
>> Second baseman. I think he has the I
00:22:16
think he has the most home Yeah, he has
00:22:19
I it says here in my notes he has the
00:22:21
most home runs ever by any second
00:22:23
baseman.
00:22:24
>> I think he has a better shot than Dill
00:22:26
Do
00:22:27
>> Don Maddingley
00:22:29
by the numbers his career was too short.
00:22:31
I just can't I can't I can't sanction
00:22:33
it. I mean he he definitely was the best
00:22:36
hitter in baseball for about two to
00:22:37
three years. um and super competitive
00:22:40
hitting you know 340s winning batting
00:22:42
title uh you know 30s one year I think
00:22:46
145 RBI's a statistic we like
00:22:48
>> I remember that year
00:22:48
>> like disparaged but is impressive great
00:22:51
great fielding first baseman but just
00:22:53
didn't have the longevity I mean he just
00:22:56
just you can't have o only if if you're
00:22:58
going to be in only on peak you have to
00:23:01
have Griffy level peak Kofax level peak
00:23:06
you can't
00:23:06
>> judge level peak
00:23:07
>> judge level peak Trout level peak. Look
00:23:09
at Trout. I mean,
00:23:10
>> yeah, Trout
00:23:11
>> go into the Hall of Fame because of 10
00:23:12
year first 10 years of his career.
00:23:14
>> That's a long peak. All right. How about
00:23:16
How about
00:23:17
>> Dale Murphy?
00:23:18
>> Also, first baseman. I don't see I I
00:23:22
don't see it happen.
00:23:22
>> All right. Gary Sheffield.
00:23:25
>> Chef Chef was quite a hitter. I think
00:23:27
his career statistics might be a little
00:23:28
higher. I don't have them in front of
00:23:29
me, but
00:23:29
>> 509 home runs.
00:23:31
>> Yeah. And there's a majesty with the 500
00:23:33
homers, right?
00:23:34
>> 1,676
00:23:35
RBI's. I think he's got the a good shot.
00:23:38
>> Okay. Fernando Venuelo.
00:23:40
>> No.
00:23:42
>> Okay. Just No. Okay. Well, those are the
00:23:44
seven contemporary. So, you could see in
00:23:47
your mind, you wouldn't be shocked if
00:23:49
Jeff Kent got in.
00:23:50
>> No, I got you.
00:23:51
>> You would be shocked if Gary Sheffield
00:23:52
got in.
00:23:53
>> Nope.
00:23:54
>> Okay.
00:23:54
>> I'm not I'm not I to listen to our
00:23:56
listeners. I'm not staring at their
00:23:58
numbers and uh I just I'm just recalling
00:24:00
what I imagine about these players. And
00:24:02
I do think Chef he had he had 500
00:24:04
homers. Um,
00:24:06
>> that that's that's a mark that almost
00:24:08
always gets you into the Hall of Fame.
00:24:10
>> Kent's leading home runs in a second
00:24:12
base position is really impressive.
00:24:14
>> Don't remember if there were uh PhD
00:24:17
performance-enhancing drugs rumors about
00:24:19
Sheffield. I think there were, which may
00:24:21
be wellkeeping,
00:24:22
>> maybe why he's where he is
00:24:24
>> because you would think someone with 500
00:24:25
home runs and almost 1,700 RBI's
00:24:28
>> going to shoot in, right?
00:24:29
>> Is in. I mean, that's in. It might be
00:24:31
third tier, but you're in. Let's talk
00:24:33
about the current ballot.
00:24:36
>> Ah, interesting.
00:24:37
>> Now, I don't think there's anyone coming
00:24:39
in on the bot on the as first timers.
00:24:41
>> Yeah. So, they're not in my list. I
00:24:43
looked at them and I just pruned them.
00:24:45
There's nobody. Nobody. Trust me,
00:24:47
nobody.
00:24:48
>> Mhm.
00:24:49
>> It's a great year for people who hanging
00:24:51
on. So,
00:24:52
>> let's talk about a couple hang Let's
00:24:53
talk about a couple hangers on and let's
00:24:55
see if you think. So, Carlos Beltran,
00:24:58
>> yeah, this is his year. He was he was
00:25:00
close last year with nobody. He's
00:25:03
getting in. Yeah, this is here.
00:25:05
>> Andrew Jones in his ninth year.
00:25:08
>> Uh, it's either this year or next year
00:25:10
he's getting in. So, I would go for this
00:25:11
year because I mean, uh, what's the
00:25:13
early reports? Do we have the early
00:25:15
reports yet? I know.
00:25:16
>> I haven't even looked. I know you like
00:25:17
to look at that. I have not.
00:25:18
>> You like to look at them? I I like to
00:25:20
look at I like to look at the Hall of
00:25:21
Fame traffic.
00:25:21
>> And that's probably it, right? Because
00:25:23
of how far the other people like The
00:25:24
next person on the list is Chase Utley.
00:25:26
He's
00:25:27
>> Chase Utley, I believe, will make it
00:25:28
eventually,
00:25:29
>> but he's in his third year at 39.8%. So
00:25:32
he's not jumping to 70.
00:25:33
>> He's not getting anywhere close quite
00:25:35
yet. Um so uh there aren't very many
00:25:38
ballots out yet right now. So um uh I I
00:25:42
right now there looks like almost
00:25:44
nothing out. So
00:25:45
>> okay. And then there's people I mean
00:25:47
some other reasonable name you know look
00:25:48
other names that aren't getting in
00:25:49
because of possibly
00:25:51
performance-enhancing drugs. Alex
00:25:52
Rodriguez, Manny Ramirez, obviously they
00:25:55
would be in based on their numbers. Andy
00:25:57
Pettit's in his eighth year won't make
00:25:59
it. Here's a guy, you know, I've never
00:26:01
quite understood. If you talk about peak
00:26:03
performance over What about King Felix,
00:26:07
>> he was great um for a short period of
00:26:10
time.
00:26:10
>> Well, I mean, extraordinary
00:26:11
>> question. Is Deg Grom in the Hall of
00:26:13
Fame?
00:26:14
>> Yes, Deg Gro's I think. I think
00:26:16
>> Well, let me see. I'm going to, you
00:26:18
know,
00:26:18
>> Deg Gro better than King Felix.
00:26:20
>> You know what? This is This is up my
00:26:22
alley because I've done this research.
00:26:23
So, I'm going to give you I I can So,
00:26:26
Deg Grom hasn't retired yet. So, um, I'm
00:26:29
going to I'm going to I'm going to pull
00:26:30
them up. I This is Yeah, as many of you
00:26:32
know, I I worked with Ryan Brill. We
00:26:34
created our Grid War metric, uh, which
00:26:36
ranks all which ranks every starting
00:26:39
pitcher based on their their quality.
00:26:41
Um, so I'm going to take a look to see
00:26:43
where Deg Grom falls. So the way our
00:26:47
metric works is it take the um the
00:26:50
geometric average of the rank in your
00:26:53
peak and your based on what?
00:26:56
>> So your rank in so the grid war so our
00:27:00
it's a long metric so your so I have a
00:27:02
annual grid war metric and you have your
00:27:04
career grid war metric um so that that's
00:27:08
how we calculate it. So um and so the so
00:27:11
if you want to just look our metric puts
00:27:13
Greg Maddox at number one. Can you just
00:27:14
speak by the way can you just just for
00:27:16
one second can you say to people the
00:27:18
advantage of taking a geometric mean as
00:27:21
opposed to an arithmetic mean here like
00:27:23
just take this why don't you just take
00:27:24
the simple average
00:27:26
>> yeah you could do that the advantage of
00:27:27
doing that is it doesn't if you have um
00:27:30
so the problem with that is that let's
00:27:32
say take Sandy Kofax Sandy Kofax is one
00:27:34
in peak um peak uh rank he's number one
00:27:40
um and uh he is 48 in career so if you
00:27:43
just took the the average you're not
00:27:45
giving enough quality enough enough kick
00:27:48
to that number one. And so in other
00:27:50
words, you you essentially it's a
00:27:52
decaying term.
00:27:53
>> And the geometric mean puts more kick
00:27:55
into peak performance.
00:27:56
>> Wait. So the geometric mean for for
00:27:58
right. So so it just it basically says
00:28:00
that if you're really really kind of
00:28:02
low, you're just low. You're not. And so
00:28:04
if you're 48 or 100, it's not that
00:28:06
different because it's it's the square
00:28:09
root metric. Um, and that's and it
00:28:11
really values a high value and and it
00:28:13
doesn't un doesn't destroy you with an
00:28:15
outlier on one of them. So, it's a
00:28:17
little bit uh I think it has slightly
00:28:19
better properties. Um, and by the way,
00:28:21
it it's extremely efficient. It it it's
00:28:24
everybody's in until you get to a point
00:28:26
where everybody's out. So, it's uh
00:28:29
>> you're telling me it has it has
00:28:30
literally a perfect set.
00:28:32
>> No, it's not perfect.
00:28:33
>> No, it has perfect. It has it has one
00:28:35
person, which we've talked about
00:28:37
repeatedly, who's not in it.
00:28:38
>> Not Kevin Brown. It's Kevin Brown. He's
00:28:40
the only exception. He's s
00:28:42
>> You're telling Oh, wait. You're telling
00:28:43
me I just want to be clear for our
00:28:44
listeners here on Wharton uh Wharton
00:28:46
Moneyball by rank order person people
00:28:49
pitchers by your metric. There literally
00:28:52
is a line I can cut where everybody
00:28:55
except for Kevin Brown is in and
00:28:56
everyone below that number is
00:28:58
>> not everyone. No, no, no, no, not every.
00:29:00
There's a couple exceptions. Um, but it
00:29:02
pretty much it goes from almost
00:29:03
everybody above except for Kevin Brown
00:29:06
and almost everybody below is not, but
00:29:08
there are a couple exceptions. So, Phil
00:29:09
Negro is uh is
00:29:11
>> and Catfish Hunter, they're not at
00:29:14
exactly at the border. Perchilling is
00:29:16
not I have to exclude those guys, those
00:29:18
guys who didn't get in because of
00:29:20
reputation. But you go down and you get
00:29:22
every single person all the way down the
00:29:24
list and to skipping over Kevin Brown.
00:29:27
You get Max Cerver, he's Sebathia, he's
00:29:29
these guys are are in now. Sherzer's in.
00:29:31
And then the one the one borderline
00:29:33
case, this is my favorite one, the
00:29:35
actual border one, the guy who sits on
00:29:36
the border who's nodding is Dave Steve.
00:29:39
>> Well, I know you've talked about Dave
00:29:41
Steve so many times. Look, you've come
00:29:44
to the conclusion that Kevin Brown is
00:29:45
more deserving than Dave Steve. I
00:29:47
>> He's more deserving. Yes.
00:29:48
>> Okay.
00:29:48
>> Yeah.
00:29:49
>> So, is there a a planet that we live on
00:29:52
where this veterans committee or
00:29:54
whatever they call current era committee
00:29:56
like Kevin Brown or Dave Steep gets in?
00:29:59
>> Absolutely. the it really depends on how
00:30:01
analytics focused they are, right?
00:30:03
Because the case for Kevin Brown and and
00:30:05
for Dave rests on having extraordinarily
00:30:09
good peak performance and nice long
00:30:11
careers. There's peak performances are
00:30:13
for both of them are majorly undervalued
00:30:16
and they're undervalued because of win
00:30:18
loss records.
00:30:20
Well, particularly with Kevin Brown,
00:30:21
these years he play he pitched for the
00:30:22
Padres's. He was a 500 slightly better
00:30:26
than 500 pitcher, but was the dominant
00:30:28
pitcher. Let's make you make a
00:30:30
prediction now. So, Eric Bradlo and his
00:30:32
three sons, I'm at Coopertown like I am.
00:30:35
It's July whatever 24th or 25th next
00:30:37
year.
00:30:38
>> You're telling me I'm going to see
00:30:40
Carlos Beltron.
00:30:42
>> You're telling me you think I'm going to
00:30:43
see Andrew Jones
00:30:45
>> mentally. You're saying and I might see
00:30:48
Jeff Kent and I might see Gary
00:30:50
Sheffield.
00:30:51
>> That's right. That's about it.
00:30:53
>> Okay.
00:30:54
>> Well, I found King Felix. By the way,
00:30:56
King Felix is number 42 ranked uh and
00:30:59
there's nobody ahead of him. There's a
00:31:02
whole lot of people ahead of him who
00:31:03
aren't in. So,
00:31:04
>> that were not in that are not in
00:31:06
>> were not in and and I don't think he's
00:31:08
going to and if you just look purely
00:31:09
based on the on the quality, I don't
00:31:11
think you're you're seeing it.
00:31:13
>> I see. I see. But uh but you know he had
00:31:16
a lot of he had a lot people sort of
00:31:17
loved him so you never know.
00:31:19
>> I got to admit if you tell me I'm going
00:31:20
to Coopertown to see Beltron Jones Kent
00:31:24
and Sheffield I'm thinking you know the
00:31:26
s you know I have tears of Hall of Fame.
00:31:28
The sum of those four is 12. I don't
00:31:30
care what you tell me. Like there's no
00:31:33
question.
00:31:33
>> No I'm just saying that doesn't excite
00:31:35
me that much because
00:31:36
>> No, it doesn't.
00:31:36
>> I mean it's not it's not that exciting.
00:31:38
Well Audi, we've talked about NBA. We've
00:31:41
talked about some NFL and rankings.
00:31:42
We've talked about some MLB. I'm glad to
00:31:44
hear again about your metric with Ryan
00:31:46
Bril. Uh this has been the first half of
00:31:48
Wharton Moneyball here on the Wharton
00:31:50
podcast network. Stay with us after the
00:31:52
break and we're going to talk to Audi
00:31:53
about uh him and his students and the re
00:31:55
research they're doing. So come and join
00:31:57
us after the break.
00:31:59
Welcome back to the second half of our
00:32:01
show here, Wharton Moneyball, the
00:32:02
Wharton podcast network. This is Eric
00:32:04
Bradler, professor of marketing,
00:32:06
statistics, and data science. I'm here
00:32:07
with my colleague, co-author, co-author
00:32:09
and friend Audi Winer, professor of
00:32:11
statistics and data science. Some
00:32:13
combination of the two of us Cade Massie
00:32:15
and Shane Jensen are here every week on
00:32:18
Wharton Moneyball. And as I mentioned at
00:32:19
the beginning of the show, one of the
00:32:21
advantages, although we love it when
00:32:23
everyone's here, one of the advantages
00:32:24
when it's just Audi and me is I
00:32:25
basically get to interview him. And so
00:32:27
we I just talked to him about some
00:32:28
statistical stuff having to do with the
00:32:30
MLB, NFL, NBA in the first half of the
00:32:33
show. Now I thought um let's pull back
00:32:36
the curtain on what AI has spent I don't
00:32:38
know at least the last 10 plus years
00:32:40
maybe 15 plus years building which I
00:32:42
consider the greatest
00:32:45
undergraduate mast's MBA PhD level
00:32:49
research opportunity for people that
00:32:51
want to apply statistics machine
00:32:53
learning now AI data science more
00:32:56
broadly uh to sports research um so AI
00:32:59
um I have no guide to you except you
00:33:02
know why don't you start by telling us
00:33:04
one of the projects that you're working
00:33:06
on now or recently that excited you and
00:33:08
we'll get to as many as we can in the
00:33:09
second half of the show.
00:33:10
>> So that's great opening just to give you
00:33:13
the listeners a little bit of
00:33:14
introductions. We have uh we have many
00:33:16
many students who are doing research in
00:33:17
in statistics, machine learning,
00:33:19
computer science, mathematics and their
00:33:21
their research is in some sort of sports
00:33:23
application which is a really um it's a
00:33:27
exciting for them because they they're
00:33:28
they're close to the edge, right? And
00:33:30
one of the way one of the way we
00:33:31
introduce our in our seminar when
00:33:33
they're research is think about
00:33:35
something that's that's that's caused
00:33:36
you to be think think about a problem.
00:33:40
What what is what what is what are you
00:33:42
curious about and and then try to get
00:33:44
the data. So we have lots and lots of
00:33:46
projects going on. Most of these are are
00:33:48
uh some of them already been published.
00:33:50
Um some of them have won prizes across
00:33:52
different sports. So we have it
00:33:54
surprisingly we don't have any
00:33:55
basketball but we have rug rugby a few
00:33:58
in in football we have tennis um we have
00:34:02
um we obviously we have baseball and we
00:34:04
have um in and question let me ask a
00:34:07
question related to that. So, one of the
00:34:09
things in the, you know, in our open
00:34:12
source publishing that we have now,
00:34:14
let's imagine there's one of our
00:34:15
listeners on Morton Moneyball that
00:34:17
wanted to either replicate or uh extend
00:34:21
some of the work that you've done with
00:34:22
your students. Like, are these data sets
00:34:24
public? And number two, if you publish
00:34:27
the paper, do you also publish code and
00:34:29
data with it? And like, you know, which
00:34:31
some journals require now, some don't.
00:34:33
like how would one of our listeners if
00:34:34
they wanted to say I want to see how my
00:34:36
skills are. I want to replicate winer
00:34:38
and brill or somebody else how would
00:34:40
they do that?
00:34:40
>> So um depends on the sport. So
00:34:42
everything we've gotten from football is
00:34:45
either public through event data uh next
00:34:48
generation stats just public event data
00:34:51
or from the NFL big data bowl. So the
00:34:53
tracking data we've used that if it's
00:34:56
soccer there is public event data but
00:34:59
our we have two soccer projects which
00:35:01
I'll happily talk about in a moment that
00:35:03
comes from our partnership with PFF so
00:35:05
pro football focus FC um you know their
00:35:08
their soccer kind of arm they have
00:35:11
collected their own tracking data using
00:35:13
video and they've let us have one season
00:35:16
and now we're about to sign on for a
00:35:18
second season. I'm not sure we'll be
00:35:20
able to make that data public, but we
00:35:21
will make code and everything related to
00:35:23
the analysis um public. So, actually
00:35:25
that's a good place for us to start
00:35:26
because we just submitted a paper. So,
00:35:29
this is probably one of our highest
00:35:30
level teams because it includes um
00:35:33
Jonathan Pippi who's a second year PhD
00:35:35
student. It includes Tion Shu who's a um
00:35:38
uh who's a master's in data science and
00:35:40
he's finishing his second year applying
00:35:42
to PhD programs right now. And it also
00:35:44
includes Paul Sabin who's our our senior
00:35:46
fellow. and they've worked on this uh
00:35:49
this this soccer tracking data to do
00:35:52
something which is uh which they call XG
00:35:54
plus. Um so most people have heard of XG
00:35:57
in soccer. XG is the expected goals
00:36:00
across a game and so that way you get
00:36:03
credit for goals that aren't goals. So
00:36:05
shots every shot has a certain
00:36:07
probability.
00:36:08
>> Is XG typically computed at the player
00:36:10
level or the team level? And well, it's
00:36:12
computed on a shot level and it's you
00:36:14
can associate them with players if you
00:36:16
like, but this is reported at the team
00:36:18
level. So, it's usually reported at the
00:36:19
game level
00:36:20
>> and it's usually an underlying metric.
00:36:22
So, if your if your team if if you you
00:36:25
can lose the game in F in XG but win the
00:36:28
game in goals
00:36:29
>> and that happens a lot. And so ex
00:36:31
>> would you since I'm trying to play as
00:36:33
host here would you relate this somewhat
00:36:36
to what in the first half of the show
00:36:37
you talked about about the Kansas City
00:36:39
Chiefs like maybe like there's some
00:36:41
underlying metric you're in here it's XG
00:36:44
plus before it was you know average
00:36:45
length of drive you know maybe that's a
00:36:49
bet there's lots of things as we all
00:36:50
know that are better indicators
00:36:51
necessarily of strength than the actual
00:36:54
randomness that happens with outcomes
00:36:56
and wins and losses
00:36:58
>> abs is exactly a parallel and this is a
00:37:00
huge leap up in soccer evaluation
00:37:03
because there are very very few goals in
00:37:06
soccer and a game can go one zero 0 with
00:37:08
a shootout and what are you going to do
00:37:10
with that but xG can be can accumulate
00:37:13
and in fact an individual player can
00:37:14
have substantial xG. Now usually in
00:37:17
sports we compare actual goals to XG and
00:37:21
we use that to attribute some special
00:37:23
quality um to the player as if they
00:37:26
consistently do that. One of the things
00:37:27
that we've learned in soccer is that
00:37:30
with maybe well certainly with one
00:37:31
exception, maybe a couple others, very
00:37:34
very few people seem to outscore their
00:37:37
xG. Their XG is just whatever is you
00:37:41
create, by the way, you create your own
00:37:43
XG. So that you get responsibility for
00:37:45
that.
00:37:45
>> By the way, how give our listeners a
00:37:48
sense of how that's computed? Is it I
00:37:51
would imagine you'll just tell me if I'm
00:37:52
wrong. I would imagine um where you're
00:37:55
shooting from is part of this, right?
00:37:57
>> The distance to the goal. Yep.
00:37:59
>> Yep. I would imagine possibly. I don't
00:38:00
know. Is the angle
00:38:02
>> angle? Yep.
00:38:03
>> Okay.
00:38:04
>> Absolutely. People were representative
00:38:06
by how much of the goal is visible,
00:38:08
right? So you take your angle and they
00:38:09
they can calculate how much of the how
00:38:12
much of of the 360 would be the goal,
00:38:14
right?
00:38:14
>> The location of the defenders.
00:38:17
>> Absolutely. They have a metric for that.
00:38:18
I mean, it's complicated and it's and
00:38:20
these are often proprietary, although we
00:38:22
built our own. um how how you actually
00:38:24
take the defenders in in to create that.
00:38:26
But basically often what they do is is
00:38:28
is essentially how much of the area is
00:38:30
open, right? So
00:38:31
>> yeah, you can literally physically say
00:38:34
if each player has a certain radius and
00:38:36
you have a certain angle and distance,
00:38:38
there's a certain amount of openness and
00:38:39
there you go. You could
00:38:41
>> So XG metrics can get very complicated
00:38:43
because you could take a look at, you
00:38:45
know, what happened immediately before
00:38:46
the shot and integrate that in as well.
00:38:48
Like uh are you almost like shooting off
00:38:50
of the dribble? Are you shooting off of
00:38:52
immediate pass? Um the Zaxis, the height
00:38:54
of the ball could be in it. Um but
00:38:57
usually they are not primary drivers are
00:39:01
defenders angle and distance and
00:39:03
distance is
00:39:04
>> look all of our listeners are saying
00:39:05
Eric when are you going to ask him
00:39:06
what's the secret sauce? So what's XG
00:39:08
plus you guys? I mean it's got to be
00:39:10
better than just XG. You call it XG plus
00:39:13
otherwise.
00:39:14
>> All right. So, just to just to close the
00:39:16
book on XG, uh Messi is the only player
00:39:19
who consistently outgo his xG and by a
00:39:22
lot. Um he's just you can see that just
00:39:25
immediately. And in fact, this past
00:39:27
Nessus or New England sports um um uh
00:39:30
conference at Harvard, he just stuck out
00:39:32
in someone's analysis and that that
00:39:34
everybody understands. Um okay, so what
00:39:36
is XG+? So what our what our what our
00:39:39
team did is they asked the simple
00:39:40
question XG is only calculated on shots
00:39:43
taken
00:39:45
and there are lots of times where a shot
00:39:47
can be taken but for variety of reasons
00:39:50
isn't and that isn't account accounted
00:39:54
for in XG.
00:39:55
>> Very interesting. So what they did is
00:39:57
they they calculated at any instant what
00:40:00
is the probability of the shot being
00:40:01
taken and then if that shot would take
00:40:05
were taken at that point that would
00:40:06
generate an XG. So you could actually
00:40:09
essentially integrate over over
00:40:11
continuous time this this this quantity
00:40:15
which is your almost and in most
00:40:17
instances the probability of a shot
00:40:18
taking is is zero. So it doesn't
00:40:20
accumulate massively, but you'll end up
00:40:22
missing all this opportunity if you only
00:40:25
look at XG as opposed to
00:40:27
>> could I in XG plus could I penalize a
00:40:30
player who should have taken a shot but
00:40:33
chooses not?
00:40:33
>> Absolutely.
00:40:34
>> That's right. And so PE so players who
00:40:37
have all this opportunity that they
00:40:39
don't do, they'll get penalized in XG
00:40:41
Plus. and players who who who take the
00:40:44
shots that they should take um you'll
00:40:47
not only see them in XG but they won't
00:40:49
get penalized.
00:40:50
>> This is now the reason I love this
00:40:53
besides it's interesting now you could
00:40:56
have two different qualities of players.
00:40:57
It's almost like false negative false
00:40:59
positive. Some players are false
00:41:02
negatives. They should be shooting and
00:41:04
they don't. And some players are false
00:41:06
positive. They shouldn't be shooting and
00:41:07
they do. So you could actually decompose
00:41:10
someone's total XG plus I assume into
00:41:14
almost like false positives and
00:41:16
negatives.
00:41:16
>> You can do a lot with a metric and it
00:41:18
has a lot of the features that you'd
00:41:20
like to see in a metric which is it
00:41:21
predicts out of sample well it
00:41:24
correlates with lot it it passes the
00:41:25
snip test. The great players show up. Uh
00:41:28
this is lots of interesting produces
00:41:30
interesting results and they're just
00:41:31
getting started with it and I don't you
00:41:32
know I didn't actually collaborate with
00:41:34
the research at all. This is one of the
00:41:35
few projects that I had basically
00:41:38
nothing to to do with it. They just did
00:41:40
it and presented. But that large measure
00:41:42
is
00:41:43
there might as you know a there are a
00:41:45
lot of people that listen to us that are
00:41:46
academics like you and me. A lot of
00:41:48
people in practice. Let me let me take
00:41:50
it from each perspective.
00:41:52
>> First where does Jonathan Paul I forget
00:41:56
the other third person's name on the
00:41:57
project.
00:41:59
>> Yeah. Where do they try to publish this?
00:42:02
>> All right. So that's interesting. Um so
00:42:04
the there are lots there's a whole bunch
00:42:06
of sports analytics journals
00:42:08
>> and there also statistics journals and
00:42:10
there in some sometimes you can publish
00:42:12
in in in sort of operations research
00:42:14
journals or econ journals or math
00:42:15
journals. Um so there are um the obvious
00:42:19
candidate for a lot of research like
00:42:21
this would be the journal of
00:42:22
quantitative analysis and analytics and
00:42:24
sports which is probably the the most
00:42:26
prestigious sportsoriented statistics
00:42:28
journal. Um there are lots there are
00:42:30
others. There's a journal of sports
00:42:31
statistics and there's there's a whole
00:42:33
bunch of sports analytics journals. Um
00:42:35
it all depends on how substantive of
00:42:37
your statistical contribution is. Um so
00:42:40
and how and and and what you're doing
00:42:42
with that. So most of our um and so you
00:42:45
could also apply to you can publish in
00:42:46
the annals of applied statistics in
00:42:48
Jazza. Um these are these are very very
00:42:51
uh um elite journals in statistics. If
00:42:54
you're making a methodological
00:42:55
contribution to statistics as well as
00:42:58
just doing a really important insight
00:42:59
into sports, one of the things that we
00:43:01
always do, in fact, if you want to read
00:43:02
about XG+ without having to wait for the
00:43:05
full publication of the manuscript, we
00:43:07
have what's called uh Wharton Research
00:43:08
Notes on our on our Wasabi web page
00:43:11
which has a a popularization of the
00:43:13
content and you and they explain what
00:43:15
they've done without getting into the
00:43:17
nitty-gritty of the maths of the method.
00:43:18
I also imagine I don't know I assume
00:43:20
they may post it like on SSRN so there
00:43:22
may actually
00:43:23
>> yeah so the archive is what we've been
00:43:25
using that's that's usually the math
00:43:27
place for that you have any what do you
00:43:30
think SSRN archive
00:43:31
>> post stuff on SSRN but either one I as
00:43:34
long you know I'm always happy to post
00:43:36
you know stuff through the review
00:43:38
process and post it there and you know
00:43:39
>> yeah so you do SSRN um so we have lots
00:43:42
of others um I um so actually
00:43:43
>> tell me about another project
00:43:45
>> yeah so why don't we talk about one that
00:43:46
I could do really quickly um one of our
00:43:48
one of our undergraduates um is is
00:43:51
working on a a a rugby paper. So uh so
00:43:55
this is really interesting. So rugby is
00:43:57
a one of the most understudied um
00:44:00
sports. So his name is Kenny Watts and
00:44:02
he's working with Jonathan Pippen. Um
00:44:04
and so you think rugby is like football,
00:44:06
right? There should be a lot of
00:44:07
analytics, right? Rugby football
00:44:10
football has analytics, rugby has
00:44:11
analytics. Well, guess what? There's
00:44:13
nothing. So what do you if I asked you
00:44:16
what do you think is the first and most
00:44:18
important analytics paper in statistics
00:44:21
goes back around 25 years um in football
00:44:23
what do you think it was
00:44:25
>> wait what's the most in in football
00:44:28
>> not in football what question in
00:44:30
football was first analyzed with
00:44:32
analytics that was really influential
00:44:35
>> um the question
00:44:38
>> yeah the paper I know of
00:44:40
>> is
00:44:42
does the kicker suck or was at the
00:44:44
distance.
00:44:45
>> That's one. You know, that's not the one
00:44:47
I was thinking of. Um I
00:44:50
>> would you go for it on fourth and one?
00:44:52
>> That's one fourth down. Right. So, uh in
00:44:54
fact, it's called Do Firms Maximize and
00:44:56
it's by Ror um and it's it's written as
00:44:59
an economics paper, but it does two
00:45:01
things. It calculates an expected points
00:45:03
model for football given down and
00:45:05
distance and and and yards to go. Um how
00:45:08
many points are expected on that drive?
00:45:10
and he uses that to decide whether or
00:45:12
not teams are going for it at the right
00:45:15
rate on fourth down and he comes to the
00:45:16
overwhelming conclusion that they don't
00:45:18
even remotely. So what Kenny and
00:45:20
Jonathan did was they said can we do
00:45:22
that for for rugby?
00:45:24
>> So explain to me I don't know even know
00:45:26
enough about the rules of rugbys and
00:45:28
stuff like if you don't get a certain
00:45:30
amount or don't score you have to give
00:45:32
the ball to the other team. I always
00:45:33
thought as as long as you guys keep
00:45:34
possessing the ball and you're going
00:45:37
forward on the field you keep the ball.
00:45:39
>> All right. So the question that they
00:45:40
asked, which I never heard of because I
00:45:42
don't know rugby, was that if a if some
00:45:44
sort of penalty happens, and I guess
00:45:46
they happen a lot.
00:45:47
>> Yeah.
00:45:47
>> The team who the offensive team, the
00:45:49
team that has the ball that's subject to
00:45:50
the subject to this penalty,
00:45:52
>> they get to choose. They can try to kick
00:45:54
what you what amounts to a field goal,
00:45:57
>> which I think is worth three points.
00:45:58
>> It is.
00:45:59
>> Or they can kick it out of bounds and
00:46:02
then they get the ball
00:46:04
essentially about 20 yards downfield
00:46:06
wherever they kicked it out of bounds.
00:46:08
And it's kind of like going for it
00:46:10
because then they can then score a
00:46:11
touchdown which is worth six and then
00:46:13
and then they have an extra point or two
00:46:15
um in in that formulation.
00:46:17
>> Just be clear, your team has a penalty.
00:46:21
I get an opportunity to either kick it
00:46:23
for three or kick it down field and then
00:46:27
of course I may end up with zero. But of
00:46:29
course I also have the ball farther down
00:46:30
the field. So I'm more like
00:46:31
>> get six or seven.
00:46:33
>> That's the same thing.
00:46:35
>> It's very similar. Seems
00:46:36
>> exactly the same thing. And the question
00:46:38
became why had no one done this before?
00:46:40
And the answer is data hasn't been
00:46:42
available and the sport's far behind.
00:46:44
And so our our team Kent Kenny is uh
00:46:47
knows rugby. Um and uh he he asked let's
00:46:50
do it. And they built an expected points
00:46:51
model um very similar to the way the ROR
00:46:55
built his first one with linear
00:46:56
regression taking every every play as a
00:47:00
having a set of coariantss and and um
00:47:02
the outcome which are highly correlated
00:47:04
and have to deal with that and
00:47:06
calculated an expense points model and
00:47:08
then he basically calculated what is the
00:47:10
optimal decision and turns out that just
00:47:13
as we've seen in almost every sport they
00:47:15
are kicking too many field goals.
00:47:17
>> That's what I assumed you were going to
00:47:18
say.
00:47:19
>> Yep. There it is. So, that was a just
00:47:22
it's a wonderful example because it's it
00:47:24
combines um a really important question
00:47:26
that people want to know the answer to.
00:47:28
>> Um we're talking about undergraduates
00:47:29
here. So, they're trying to flesh out
00:47:32
techniques that are pretty standard at
00:47:33
this point and they apply them in a new
00:47:35
situation and they tell us something
00:47:36
that we almost expected to happen, but
00:47:39
now it actually quantifies it.
00:47:42
Can you give us a sense of like how many
00:47:45
points or how much win probability a
00:47:48
team is giving up? Like I always like
00:47:50
you know since I try to be the effect
00:47:51
size guy here.
00:47:53
>> So you know teams aren't optimal. Okay.
00:47:56
But are they like you've pointed this
00:47:57
out even with your work with Ryan Bril
00:47:59
like people don't go for it on forth not
00:48:02
necessarily always the right time but
00:48:04
sometimes it's not irrational depending
00:48:07
like you could come up with a risk
00:48:08
aversion story or you come up with an
00:48:10
uncertainty story that explains it. Is
00:48:13
it that you know sure they're kicking
00:48:15
too many field goals but they're
00:48:18
basically not losing much win
00:48:20
probability or maybe they are. I mean
00:48:21
how big an effect are we talking about?
00:48:23
that's got to be something on their
00:48:25
docket. One of the things that when I
00:48:26
deal with students is that I teach them
00:48:29
pretty aggressively that it's important
00:48:31
to finish and not do everything. And
00:48:35
>> oh, I thought you meant finish by a
00:48:36
different thing. So, let me I'm going to
00:48:37
just This is not a story.
00:48:39
>> What I mean is that they they they have
00:48:40
to they need to finish a b you you know
00:48:44
this. It's never done. Right.
00:48:45
>> Right. It's never done.
00:48:46
>> Never done everything. No, but the story
00:48:48
I was going to point out is why, you
00:48:50
know, when I while I publish obviously
00:48:52
in both statistics and marketing, in
00:48:54
some ways marketing's just different,
00:48:56
not harder or easier, just different
00:48:58
because let's say you built this
00:49:00
statistical model for something. Let's
00:49:03
say kicking field goals in rugby or
00:49:05
kicking it. Then someone would you could
00:49:08
never publish that. You'd have to go to
00:49:11
the next step. And so what does this
00:49:13
mean for firm performance? What does
00:49:14
this mean for winning? like they will
00:49:17
never let you like I I thought you meant
00:49:19
finishing by and so tell me what this
00:49:22
means as opposed to you have to draw the
00:49:24
line somewhere and say you've built a
00:49:26
good predictive model you've answered an
00:49:28
interesting substantive question you're
00:49:30
not yet you know you could extend it and
00:49:32
say oh what does it affect when
00:49:33
probability but at some point the
00:49:35
project just ends if you're interested
00:49:36
in that answer it again
00:49:38
>> that's a great question because actually
00:49:39
the the getting an effect size I think
00:49:41
is something that needs to be done
00:49:42
before they submit that that's too
00:49:46
important a question. So they built a
00:49:49
model, they have a a conclusion, they're
00:49:51
not they're not aggressive enough, and
00:49:52
now you have to kind of turn it into how
00:49:54
many points you're giving up or
00:49:56
something and which probably then turns
00:49:58
into wins pretty easily. So that piece
00:50:00
is is I think is important. But you
00:50:02
know, these models are going to be
00:50:04
incomplete. They're models, right?
00:50:06
You're not going to be bringing every
00:50:07
factor in. And um you could be look
00:50:09
doing looking at win probability models
00:50:11
instead of expected points models. And
00:50:13
there's many ways to do these things.
00:50:15
How you going to be treating the the the
00:50:17
how you dealing with the independence?
00:50:18
Are you how you dealing with the
00:50:19
standard errors? There's so many deep
00:50:21
questions that you could ask which
00:50:23
eventually should get get answered but
00:50:25
can't do them all in the first round.
00:50:26
>> By the way, Audi, I would have lost
00:50:27
about a million dollars. And let me tell
00:50:29
you why.
00:50:31
If you had told me that you and I were
00:50:33
going to talk about
00:50:36
related research, no. No. The first two
00:50:38
you would bring up would be soccer and
00:50:40
rugby. I'd be like, I'll give you a
00:50:44
million to one odds. So maybe you could
00:50:47
tell us about one that deals with the
00:50:49
students that I have a little more
00:50:50
knowledge. No, no, I'm not saying
00:50:52
there's anything
00:50:54
>> but
00:50:56
we have one in tennis which you I'll
00:50:57
just
00:50:58
>> I want to I love tennis. So get to
00:50:59
tennis. But before I do that, I want to
00:51:01
point out to our listeners this what
00:51:02
Audi's pointing out is important which
00:51:04
is I've always said I joke about it with
00:51:07
my friends. There's only one thing we as
00:51:09
academics do really well and that's we
00:51:13
recognize the isomorphism between
00:51:15
problems. So if you're not a rugby fan
00:51:18
and you're not a soccer fan, who cares?
00:51:20
What Audi's talking about is applicable
00:51:23
to so many other sports or so many other
00:51:25
problems. This is why sports is
00:51:27
wonderful as a testing ground for
00:51:29
statistical methods and learning because
00:51:32
it's, you know, you can change the name
00:51:35
from, you know, kicking field goals and
00:51:38
scoring touchdowns in one sport to
00:51:40
another or, you know, you have two
00:51:42
options A and B and one is more certain
00:51:44
but has lower highend. You know, that's
00:51:46
what I love about sports is that you can
00:51:49
learn a lot about business and business
00:51:51
decision making through sports type
00:51:53
models.
00:51:54
>> All right. So you're going to tell us
00:51:55
about tennis.
00:51:56
>> Uh so so tennis actually um this has
00:51:58
came up. These are this is a two two
00:52:00
students um did this um they uh they put
00:52:03
this together um and it's really it's um
00:52:06
and they they presented it and at
00:52:08
Carnegie Melon um so let me just tell
00:52:10
you who they are. It's I win Amita and
00:52:13
Audrey um Rita we call her and uh so I
00:52:16
win Rita and Audrey um they uh they
00:52:18
actually won the student competition
00:52:20
award uh for best paper at the Carnegie
00:52:22
Melon Sports Analytics Conference.
00:52:23
Great. And it was uh uh on tennis and
00:52:26
the feature of tennis was is um they
00:52:28
wanted to rank serves and the problem
00:52:30
the the problem with most ser service
00:52:32
ranks is that they don't actually
00:52:36
measure the service quality as distinct
00:52:39
from the player quality, right? Because
00:52:41
you just look how many service points
00:52:42
you're winning. Now you can look at
00:52:44
aces. Of course that's only one feature
00:52:46
of service. The other feature is getting
00:52:48
the the opponent off balance enough to
00:52:50
win the the the point fairly quickly.
00:52:52
but other ast as aspects of your play
00:52:55
kind of get involved. And so it's
00:52:57
essentially a decomposition of the um
00:52:59
the ser it takes you if you're winning a
00:53:02
lot of service points, it breaks it down
00:53:03
and to how much of that's due to the
00:53:05
quality of your serve and how much is
00:53:06
due to the quality of your player. Um
00:53:08
and it gets a
00:53:09
>> person serving or the opponent
00:53:12
>> uh both of them and adjust for opponent
00:53:13
as well. That was a key key feature. You
00:53:15
got you got for opponent.
00:53:17
>> So does it have a ranking of player
00:53:20
>> rank serves and it and it and it's
00:53:22
players
00:53:23
>> uh current I think it has some of that
00:53:24
current players I I'll see in there I
00:53:26
can look through their players and see
00:53:28
who they rank as excellent but I'm sure
00:53:30
you would enjoy that right so that's
00:53:31
>> I would I would enjoy knowing but it's
00:53:34
it is good to know because it's one of
00:53:35
those things where like simple metrics I
00:53:39
have a prediction of who the best server
00:53:41
is but it's not going to be someone if
00:53:43
if there is a table is there a table
00:53:45
should I guess somebody
00:53:46
>> uh I don't actually have it so you can
00:53:48
you can guess it for next time
00:53:50
>> I'm I'm gonna guess it's a lessernown
00:53:53
player who is 6'10,
00:53:56
>> right?
00:53:57
>> His name's Riley Opelka.
00:54:00
>> I'm going to guess that he and you
00:54:02
remember from our childhood Rosco
00:54:04
Tanner.
00:54:05
>> Well, it was Rosco Tanner. Yeah.
00:54:06
>> No, no. I'm just saying Rosco
00:54:08
>> 50 mph. Rosco Tanner.
00:54:10
>> Rosco Tanner could serve. Rosco Tanner
00:54:12
could really serve. That was the only
00:54:14
great.
00:54:17
But either way, that's a fascinating
00:54:19
project. Um, so yeah, and that and those
00:54:22
had great success and and it's funny
00:54:23
because I mean it's actually really
00:54:25
really nice to see uh these students all
00:54:27
work with me or they were all Moneyball
00:54:29
Academy TAs in part of our lab this
00:54:31
summer. Um, and that's where they did
00:54:33
their work. Um, I can finish up with a
00:54:34
few others. We have we have two
00:54:37
actually, you know, pretty good ones in
00:54:38
football.
00:54:38
>> Two more. Two more would be great.
00:54:40
>> Yeah. Well, so one one of them in
00:54:42
football. We have we have a soccer, we
00:54:43
have football, we have a baseball one
00:54:45
that's just getting started that that uh
00:54:47
we just submitted to Saber
00:54:49
>> question in the baseball one
00:54:50
>> and the baseball question actually it um
00:54:53
so it has to do with my uh historical
00:54:56
admiration of the RBI and and um so the
00:55:00
RBI statistic would know everyone likes
00:55:01
to make fun of because it's contextual,
00:55:03
right? It's it's so dependent on on the
00:55:06
the settings the opportunities you get.
00:55:09
So that that that just essentially is
00:55:12
just asking you to adjust for context.
00:55:16
So the question we asked was what are
00:55:18
your RBI's above expected?
00:55:22
>> That's very reasonable.
00:55:23
>> I I'd love to know like Lou Garri's 180
00:55:26
or
00:55:28
>> Lou Garrick the thing this is what
00:55:29
you're pointing out. The greatest season
00:55:32
I've ever thought about was when Lou
00:55:34
Garri hitund this 1927 the year Ruth hit
00:55:37
60. So, we know he came up at least 60
00:55:39
times with nobody on base.
00:55:41
>> He had 175 RBs. I think Ruth had 160
00:55:45
something. So, how many guys were
00:55:47
goddamn left on base by the time Garrick
00:55:50
got up there? And he still hit 175, I
00:55:53
think, is the number.
00:55:54
>> So, to me, that that's an obvious, you
00:55:56
know, you always ask for, isn't that the
00:55:57
one of the most obvious unanswered
00:55:59
questions in baseball? If you report
00:56:01
RBI, you should also be reporting RBI's
00:56:03
above expected immediately. And then you
00:56:06
can also um and this came from our
00:56:08
conversations with some of the analysts
00:56:09
at the Phillies. They want to know what
00:56:11
your RBI's above expected is after
00:56:14
controlling for the quality of player
00:56:16
that you are. So we know that right so a
00:56:19
lot of that RBI's above expected could
00:56:21
be due to the fact that you are a great
00:56:22
player. So that's that's that. So they
00:56:25
want to decompose the RBI above expected
00:56:28
into two pieces. one which is what you'd
00:56:31
expect to get given the quality that you
00:56:33
are and then the additional part which
00:56:35
was essentially be the luck or the um
00:56:38
you can call it luck or skill to drive
00:56:40
in those extra RBI or pressure
00:56:43
performance
00:56:45
expected part I can imagine a bunch of I
00:56:47
can imagine a statistical model but I
00:56:49
can also imagine just binning players
00:56:51
into quality buckets or tersiles let's
00:56:54
say and then say of the players in this
00:56:57
tersile based on some metric or
00:56:59
something or some per, you know, skill
00:57:01
estimate. Let's look at how many RBIs
00:57:04
above expected they have for their
00:57:06
decile. That that would be one way to do
00:57:08
it. Would you favor that over some sort
00:57:11
of let's call it parametric typo or some
00:57:13
sort of model?
00:57:14
>> Well, I I generally my my general
00:57:17
tendency is to use a parametric model
00:57:19
and then and essentially predict your
00:57:21
RBI's given your context and your say
00:57:24
WOBA. That's a simple way to do that. I
00:57:26
like your bucket. In fact, Eric, I'm I'm
00:57:28
gonna ask you. You want to be a you want
00:57:29
to be the faculty adviser to our to our
00:57:31
little team? They'd certainly love to
00:57:33
have you.
00:57:33
>> I'd love it. Let's do it.
00:57:35
>> So, we have we have we have two
00:57:36
undergraduates working on it right now.
00:57:38
Talia and uh Lev are are sophomore and
00:57:41
freshman. In fact, Leev um took
00:57:43
Moneyball Academy and he's now here as a
00:57:45
freshman and is already jumping in to do
00:57:47
research. Um I'll finish off with a
00:57:49
football um
00:57:51
in Harvard um and did a great job. They
00:57:54
did something that everyone is
00:57:55
interested in. You know, we we look at
00:57:57
wind pass uh you know that block block,
00:57:59
you know, brush win rate, right? That's
00:58:01
a standard statistic that is get
00:58:02
calculated. But guess what it's not
00:58:04
doing? It's not adjusting for um
00:58:08
quality, the quality of your opponent.
00:58:12
And that is something they want to do
00:58:14
and they so this team has tried to
00:58:16
adjust for the quality of the opponent
00:58:18
and the context and the actual what's at
00:58:20
stake, right? So failing to block in a
00:58:22
situation that doesn't really matter is
00:58:24
much less important or damaging than
00:58:26
failing to block or succeeding to block
00:58:28
and that leads to a pressure or sack. So
00:58:31
essentially they want to do is a much
00:58:32
more sophisticated evaluation of both
00:58:35
the blockers and the and the and the um
00:58:37
the rushers and they come out with great
00:58:38
great results and it's much better than
00:58:40
just you know success rate block you
00:58:42
know black uh block pass success rush
00:58:46
win rates. Well, the reason I love all
00:58:48
these stories is first of all, um, it's
00:58:50
great that you're giving back to our
00:58:52
students and giving them an opportunity
00:58:53
to do research. That's obviously number
00:58:55
one. That's great. It's great. And a lot
00:58:56
of these people, as you said, may go on
00:58:58
for research jobs in industry or PhDs,
00:59:01
etc.
00:59:01
>> Well, they're applying for statistics,
00:59:03
PhD departments,
00:59:05
>> I mean, that's that's incredible. Um the
00:59:08
second thing is it shows people that um
00:59:11
you know you don't need a PhD to do
00:59:15
interesting research too. And so that
00:59:18
part is is interesting. That part is is
00:59:21
really interesting as well. Well, this
00:59:23
has been one hour here of Wharton
00:59:25
Moneyball on the Wharton podcast network
00:59:27
on behalf of myself, Eric Bradlo, my
00:59:28
colleague, co-author and friend Audi
00:59:30
Winer uh in Absentia, Cade Massie and
00:59:33
Shane Jensen who'll be back next week.
00:59:35
Thanks to our associate producer and
00:59:37
sound engineer, Dion Simpkins. Thanks to
00:59:39
our producer and if you like our big
00:59:42
bosses, Marissa Rena and D Patel. Thank
00:59:45
you for joining us here on Morton
00:59:46
Moneyball between now and next week.
00:59:48
Enjoy your sports. Enjoy your
00:59:49
statistics. We'll see you back here on
00:59:51
the Wharton Podcast Network.
