00:00:01
Welcome, welcome to Wharton Moneyball. Welcome to a full hour of sports analytics here on the Wharton podcast
00:00:08
network. This is Kade Massie hosting this week with my two longtime collaborators, friends, colleagues here
00:00:14
at Wharton, Eric Bradlo, and Audi Winer. Our fourth co-host, Shane Jensen, is out
00:00:19
this week. He's teaching. He'll be back. He and Audi are juggling teaching
00:00:22
schedules this quarter, but we've got three of us in here for the duration. We
00:00:27
are coming up on 12 years. We've been doing this virtually every week uh for
00:00:33
12 years. Some combination of us in here. We're going to hit our anniversary
00:00:38
in March. We're recording on Tuesday afternoon as we typically do. Show will
00:00:41
go up on Wednesday. We are going to run our regular format this week, which means in the first half the show, bring
00:00:49
a guest in here who's already here. And second half, we'll do open lines, kick
00:00:54
things around a little bit, see what the boys have been thinking about over the last week. Guest this week, longtime
00:01:01
friend of the show, old friend of the show. I don't mean old man, I mean oldtime friend, Ben Alamar. You probably
00:01:09
know him from his involvement in sports analytics. Really one of the first guys in working in the NBA. One must have
00:01:16
been one of the very first full-time guys in the NBA. He's worked with the Spurs, the Cavs. It's been a good
00:01:22
stretch at ESPN working with people like Dean Oliver building out really kind of
00:01:26
building out the sports analytics prof platform there at ESPN internally and externally that has had a profound
00:01:33
influence on the world of sports. He teaches in various places. He's teaching
00:01:37
right now at the University of Texas. He's taught in universities around the
00:01:40
country teaching a sports analytics class. Has a has a book out on sports analytics done well enough to be in
00:01:46
second edition. and Ben Alamar, good to see you. Thanks for making time for us. >> Oh, pleasure to be here again. Uh, it's
00:01:53
been uh, you know, as I say, old oldtime uh, participant here, but always glad to
00:01:59
join in the discussion. >> Thanks, man. Good to see you. Um, fellas, uh, Audi is going to be up at
00:02:04
the MIT conference in a couple of weeks, the Sloan MIT Sloan Sports Analytics Conference, 20th annual.
00:02:11
Those guys run up there every year. Audi will be up there. Ben's Ben's a a
00:02:15
frequent host of some platform event session. Ben's got a session this spring
00:02:20
on academic collaborations, I believe, or academic research. So, uh, do y'all
00:02:24
need to cross paths while they're running around Boston? >> I hope I'll be able to attend.
00:02:29
>> Yeah, absolutely. >> We're slowly tearing our attention away from the football world and casting
00:02:34
around for other things to um to to attend to. Um Audi Audi of course is already excited about whatever meager
00:02:42
offerings baseball has right now. They're they're up and running, but really there are two sports underway and
00:02:48
we're trying to get deeper on both of them. Setting aside for the moment Olympics, um basketball and hockey and
00:02:54
you you've been you're you're you're good across all sports, but I think
00:02:57
you're especially deep in basketball. So, can you help us get to up to speed
00:03:01
on the NBA and what do you as you look around, as you dig in, as you pay attention, what do you think's
00:03:07
interested in the NBA this year? >> Yeah, I mean, the the NBA has been really pretty exciting this year. I
00:03:13
think, you know, there's always some some issues folks have and uh with with
00:03:18
tanking and such and we'll get to that at some point, but um I think what's
00:03:22
great about this year is we're seeing, you know, for one, the Spurs, the rise
00:03:26
of women really starting to fulfill the promise that everybody had expected. We're seeing some real surprises as well
00:03:34
that you have these the young Detroit team um really leading the league now in a lot of ways and um uh in a great
00:03:42
position to make a a strong run in in the postseason. And you also have these this other surprise the Boston Celtics
00:03:49
who you know during the offseason they were you know tearing their team down you know sent a lot of salary off a lot
00:03:55
of good players off um but have been able to put together a a really strong team that is competing well in in the
00:04:03
east uh which not as strong as the west but the still uh they've been against
00:04:08
New York and and Detroit they've been really good uh despite not having uh Jason Tatum uh and the idea that Tatum
00:04:15
is getting ready to perhaps join them for a playoff run makes that a really scary team when you know nobody was
00:04:22
really counting on the Celtics. They saw them as much like the Pacers sort of taking a gap year here while they their
00:04:27
their best player was on the bench. Uh >> then can I jump in there real quick? Um
00:04:32
the four franchises that you just pointed out I you pointed out San Antonio actually maybe you didn't point
00:04:38
out Oklahoma City but I can't help but think about Oklahoma City. So you didn't. So, San Antonio,
00:04:44
Detroit, and Boston. San Antonio and Boston, I think anybody would call two of the bestrun franchises in the NBA.
00:04:51
Famously good front offices. And of course, OKC is floating around in the background as well. I don't know
00:04:56
anything about Detroit's front office. Are they thought of in the are they thought of as an upand cominging? What
00:05:00
is it that's going on around there that allowed them to kind of surprise with a
00:05:04
young team here? Yeah. So they're so they a couple years ago uh they brought
00:05:09
in uh Tjan Langden and um uh uh Michael Blackstone. Uh both the guys have been they they were working together in New
00:05:18
Orleans before. Uh I worked with with Blackstone in Cleveland. Um he's really
00:05:22
sort of a negotiation salary cap expert. Trejan's just been um you know come in
00:05:28
and really worked his way up in the league in the front office is very well respected generally speaking and they've
00:05:33
done you know they they had some young talent there already when they got there uh that the old regime had put in place
00:05:39
but they also had a lot of you know contracts that they needed to move and and turn into valuable assets which they
00:05:46
they they've done a great job of combining sort of creating uh a platform for their young players to uh
00:05:52
demonstrate that they they can uh play while uh not um you know uh by the same time adding some some valuable pieces
00:06:01
that can help them uh you know win games as those young guys develop and get a little bit more experience. So, uh, I
00:06:07
think they've done a really nice job over the last couple years, uh, you know, transitioning that team from what
00:06:11
it was, which was, you know, u sort of mired in, uh, not being able to move forward too much. Uh, created some
00:06:19
flexibility for themselves and really has, um, you know, obviously, uh, put together a really talented roster that
00:06:25
that's doing a great job. >> Cool to hear. Um, Trajan Lane done uh Duke Colomb. Yeah, see the guy played
00:06:32
Alaska, drifted down from Alaska, played ball at Duke for a while. >> That's perfect.
00:06:37
>> Interesting. I I don't Has anyone ever tried to map the influence from various
00:06:42
collegiate programs into the NBA? Some They've got a number of former players
00:06:46
and successful front offices, don't they? >> Yeah, Duke. Duke certainly does. Um, but
00:06:52
uh it's it's hard to track. So, the front office guys are so varied in in in
00:06:56
their experiences. I mean, you take San Sam Prey, he is, you know, from D3 school at uh and and
00:07:03
>> you know, has obviously is really thought of as the top the best run franchise right now in the league. So,
00:07:09
>> Right. Right. Was he I assume he was there when you were there back in the
00:07:13
early 2000s. So, how early in Sam's regime was your time with him? >> Well, so Sam got hired in SE you know,
00:07:20
it was the SuperSonics at the time. uh he got hired um and then you know his his first decision was to draft Kevin
00:07:27
Durant. Um and then that he hired me. So it was a really big summer for the uh we we we got going. So yeah, I started
00:07:37
working with Sam right after his first draft. Um >> okay, >> you know, suffered through a couple of
00:07:43
tough years as we were building that team, but um turned it turned it around with some brilliant draft work from Sam.
00:07:49
This >> this is a a big a bigger question than we have time for, but I'm curious to get
00:07:54
just a first taste from you on this. What do you think Sam had to learn? Like what are some of the fir what what's one
00:07:59
of the first things he had to learn as he moved into that position? Presumably he kept on learning for a long time.
00:08:04
He's gotten so good at it, but what's something that you saw him work through
00:08:06
and you think he learned early on in his time? >> Um so I think that you know one of the
00:08:12
things was um sort of how to you know use it. He had come from the Spurs where you they were one of the early teams
00:08:19
that had done anything in this in the area of data analytics u and he knew that he really wanted it but he didn't
00:08:25
really um have a you know full grasp of what was possible or how to incorporate it you know really into decision-m uh
00:08:32
you know how to trust and then but you know he's he's so smart he's really good
00:08:37
at just asking questions and asking questions which is sort of the key piece that you need in that position to to to
00:08:43
grow and learn how to utilize things well and so he really developed from sort of like, all right, here's a
00:08:48
spreadsheet that we can provide for you every, you know, week that's updated to
00:08:52
really understanding how to build a system within the team, uh, and not just a one-off guy, but like a how do we have
00:09:01
this all this information really permeate through uh, everything that we're we're working on.
00:09:06
>> Yeah, that's so interesting. I think Eric was trying to jump in. >> No, no. I wanted to go back just for a
00:09:10
second to yourself comment. I'm just wondering, Ben, if you think that >> um in some sense, if they were in the
00:09:17
West, >> I think you would agree the Celtics would not be the favorites in the West.
00:09:22
>> Oh, no. >> But the fact that they're in the East also, which gives them, let's call it, a
00:09:26
round or two to get Tatum back in the flow. >> Yeah. >> You could make an argument that they
00:09:33
could be higher than Let me make an argument. You just tell me I'm wrong. >> Sure.
00:09:37
>> They're high because of the weakness of the East. >> Yeah. the openness of the East. They're
00:09:42
higher than any team in the West because in the West it could be the Spurs, could
00:09:45
be the Nuggets, it could be the Thunder, >> but out of the East, they might be the
00:09:50
favorite with Tatum back and so they have a higher probability of making it to the finals even though they have a
00:09:56
lower probability of of winning the finals. What What do you is if you had to guess right now, would they be your
00:10:02
favorite based on that argument or is that not right? >> Uh I I I I think it's it's it's a it's a
00:10:07
fair argument. I'm not sure that they're the total favorites just because we
00:10:10
don't know what Tatum is like. Is he coming back and how he meshes with how the team is playing this season? Uh, you
00:10:17
know, Jaylen Brown there has been the guy making that team work. And while Brown and Tatum have obviously played
00:10:24
together a lot, not in this particular situation and not in the case where Jaylen's gotten the the taste of really
00:10:30
being the lead of the team. And so how those guys as Tatum works his way back, you know, he's not gonna he's gonna be
00:10:39
he's way back into basketball shape, got, you know, get everything going again. Um having that their how they
00:10:46
work together in this new uh you know, takes some time. Um I mean we just when players get traded that it
00:10:56
takes time for them to settle into the new environment. Um, I think Tatum will take some time to settle into these this
00:11:02
new Celtics environment in similar ways. Um, if and they figure that stuff out quickly, then yeah, they're probably
00:11:10
favorites. Um, but, uh, you know, Detroit and New York are are are still really good. They will give them a
00:11:17
strong run for money in the playoffs. But um certainly not the stumbling blocks that uh exists in the west- w
00:11:23
with OKC San Antonio Denver as >> Ben let's talk about the west. You mentioned women coming on the team
00:11:30
really coming on. OKC was such a force last year and they seem just stocked for years to come. How does that rivalry
00:11:37
shape up and and why is it if if San Antonio is kind of surprisingly competitive in that rivalry? Why is
00:11:44
that? Yeah, it's it's really interesting because the um the Spurs are are are you
00:11:49
know look like Thunder killers right now. They are the ones who are have figured out how to beat them. Uh they
00:11:56
you know they've won they've handed the Thunder four of their 13 losses uh this
00:12:00
season. Um you know so clearly they've got something going there. Um and so you
00:12:07
know it you can see the impact in a couple of ways. Um, first, you know, on the offense, like the Thunder overall
00:12:13
have the fourth ranked offense in the league when in terms of efficiency. Um, but against the Spurs, they would be
00:12:20
like the 26th ranked offense. You know, they, you know, dropping from 120 points
00:12:24
per possession to 112. Um, they they shoot less efficiently. Uh, they get fewer fouls. They just they don't
00:12:31
>> Okay, but does doesn't everybody look worse against the Spurs? Are they not
00:12:35
just that that >> the Thunder are so good that I mean it's a dramatic drop off, right? Like they
00:12:41
they they I mean this is a team that we expect to uh you know potentially be a perennial champion, you know, and really
00:12:49
put together a a run here. And you know, they're only you know scoring like again
00:12:54
like the um the 26th best offense in the league when they're playing the Spurs.
00:12:59
>> Yeah. They're not regressing to the mean. They're reg they're regressing
00:13:02
beyond the mean to the bottom. It's polar opposite what's going on there.
00:13:05
>> Let me ask him, Ben, how much weight do you put in the regular season record of
00:13:09
the Thunder right now? Actually, by the way, they have 14 losses. I didn't even
00:13:12
realize this. The Pistons now have the best record in basketball. I actually did not know that the Thunder had uh
00:13:18
passed them for the best record. Is it does it mean anything to the Thunder? I mean, yeah, they'd like to hold off the
00:13:24
Spurs. Trust me, they'd love to have homecourt advantage in that round. Yeah,
00:13:27
that does mean something. But whether they win 60 games, 63 games, 58 games, you know, does it really mean anything?
00:13:36
>> Um, it's it's I think it it depends a little bit. I think as early on the
00:13:41
season when this the Thunder were just like they were on pace for, you know, 70 plus games.
00:13:46
>> Yeah. >> Like then then it was like, well against that, by the way,
00:13:51
>> what >> I bet it against that mean is a powerful force. >> It is. They're really good though. Um,
00:13:59
but as like at that point there there is a you know uh a decision you have to make as a team like do we care about
00:14:06
this? Should we fundamentally care about doing this? Um you know a few years ago
00:14:10
the Warriors decided that they did and they went for it uh hard. Um the this year the the Thunder you know they've
00:14:17
>> I think they've had enough injuries and things that uh you guys have missed
00:14:21
games where it became clear like that's not going to be uh going to happen for
00:14:26
them this year. They're not going after a win record this year. So, generally,
00:14:29
no, they want the they want the one seat. They'll they'll play for it. They
00:14:32
don't they don't like the Spurs. So, they there is a you know, a a rivalry
00:14:37
there was fun to see. It makes the games, you know, gives it a little extra edge because they clearly don't like
00:14:42
each other. Um so, you know, in general, yeah, they don't care between that's not an issue. One seed and Spurs
00:14:52
do matter to them. So I on the other end of the spectrum, we're hearing talk about tanking. I
00:15:00
think the NBA just fined. Was it Utah? Somebody got fined for taking out players in a competitive game.
00:15:07
>> It it you know, it's been a few years since everybody was up in arms about
00:15:11
tanking with the NBA, but it seems to be back kind of in force. What's going on?
00:15:15
And what do you think, Ben, could be done about it, if anything? >> Yeah. So, I think this what two two
00:15:21
dynamics have sort of led to the the the level of tang we're seeing right now,
00:15:25
which is, you know, the Jazz not playing their starters in the fourth quarter. Um, you know, like that's that's that's
00:15:32
beyond some of the things that we've seen in the past in terms of tang like that's pretty uh extreme uh situation.
00:15:38
>> Well, Ben, let me jump in real quickly. This is maybe cheap, but I don't know
00:15:43
what Danny Ang's role is out there in Utah. You probably have. Can you tell us
00:15:46
something about the role? When Danny Ames was a player, he would take any edge he could find. It wouldn't be
00:15:51
surprising at his organization would be Am I Is that unfair to say? Probably unfair.
00:15:56
>> Oh, I just want to interject. As a Knicks fan growing up, you put me with
00:15:59
Larry Bird, Kevin McCale, Parish, and DJ, and I'm a winner, too. Sorry. >> Oh, no.
00:16:06
>> I'm not I'm not And so, I'm not giving Danny A credit for anything on that. But
00:16:11
Ben, please go ahead. I Danny H like he's he has regardless of how you think
00:16:16
about him as a player as an executive he what he did in Boston >> you cut you cut out there for just a
00:16:23
second Danny A as executive what >> as executive has been incredibly successful you know he built the team in
00:16:28
Boston did an excellent job you know bringing talent in there has won championships as an executive um moved
00:16:35
to you know to Utah now has this a similar role there uh you know sort of re trying to build that team and he is
00:16:42
you believes and and that it's hard for Utah to acquire players other than through
00:16:47
the draft to high level top level players other than through the draft and so he has to feel like um you know we
00:16:54
have to do it through the draft and tang is the way to do it and so >> is that is that because people won't
00:16:58
come out to Utah is that >> that's that's that's the you know one of
00:17:01
the concerns is that you free agents don't look at at Salt Lake City as a uh
00:17:06
destination um both from desiraability of the city and also just from the the the the sponsorship opportunities are
00:17:13
not the same as they are in places like LA and New York. So small markets are always at this kind of disadvantage. Um
00:17:19
and Utah I think is sort of an extra disadvantage >> but um so >> I mean it sounds like you can't win at
00:17:27
all out there. I mean winning only through the I mean how long do you get to hold on to a player? Usually they're
00:17:32
not >> they have to free a player once you have a player through the the way that you
00:17:36
know through a few cont through the first contract you're the one who gets to pay them the most afterwards. So you
00:17:41
have a chance to to what OKC is doing right now. Uh I mean that's how they they they through the through a one very
00:17:48
smart trade which helped them accumulate a lot of draft assets and some other trades even more draft assets. They've
00:17:53
you know really built a uh a juggernaut without free agents. Um they've done it
00:17:59
through you know they acquired Shay through a trade. Everything else is is draft. Um,
00:18:04
>> but everybody would have said, everybody thought Prey was greatly disadvantaged
00:18:07
operating out of Oklahoma City for the same, >> but also they ended up with Shay. I
00:18:10
mean, almost that's crazy, right? >> Yeah. Yeah. They the the Clippers were
00:18:16
put in a position by by Quai Leonard where they basically had to acquire Paul George if they wanted Quai. And so uh
00:18:24
you know Sam knew that he had that the leverage to do that and and he insisted on including Shay in that trade. now
00:18:30
should clearly exceed the expectations of everybody. Uh but yeah, he's been you
00:18:35
know uh like that that that was been a real thing. Um but overall in terms of tanking like you know we've had a
00:18:42
flattening of the the odds in the in the lottery. So you know that if you have the worst record in the the league you
00:18:48
know you don't it's hard your odds of getting the number one pick are are are
00:18:51
less than they used to be. And so you so more teams have, you know, in that, you
00:18:56
know, one to six range have a shot at a top four pick, uh, a reasonable shot. And this particular draft coming up is
00:19:03
thought to be one of a a really, really strong draft with at least three guys who are thought to be significant, you
00:19:10
know, uh, franchise level players in in the future. And so you have those two dynamics coming together to really push
00:19:17
teams incentives to want to tank and make sure they're in that bottom six, bottom eight at least to have a
00:19:24
reasonable shot at keeping at getting one of those guys uh in the draft. Ben, has there been any evidence that you've
00:19:31
seen or looked at that suggests that um like I understand like you know, let's
00:19:36
say obviously LeBron James was number one and there were a lot of great players number one, but also there were
00:19:41
some number one players that didn't work out as well. Like >> how much information is there really to
00:19:47
say, well, number three is so much better than number seven or eight. Is there any evidence to suggest that's
00:19:53
really true? >> Oh, yeah, for sure. I mean we you can look at the the at draft you know uh you
00:19:58
know uh uh win probability added over by draft's position and there's a real
00:20:03
curve there's a definitely a decline curve now that's that how steep we that
00:20:10
curve is varies significantly by the draft um because as you point out like I was in Cleveland when we drafted Anthony
00:20:17
Bennett number one overall like and if you don't know who Anthony Bennett is
00:20:21
it's not your fault it like that that happens like there are busts. It's not a but um some drafts are
00:20:29
stocked with players that everybody believes great and some most of them a lot often turn out to be great.
00:20:37
Sometimes it's stocked with players that people think are going to be great and
00:20:40
they're fine. Andrew Wiggins is a great example of that. Um, and then you have
00:20:45
guys that we totally miss on, uh, like Giannis, you know, was not a, you know, when we were thinking in that, you know,
00:20:51
2013 14 draft, um, you know, uh, I guess 2013 draft, Giannis wasn't on our board
00:20:58
for one of the top five with the top pick. Um, because we just didn't have the information we needed to to to
00:21:04
believe he was at that level. But, you know, obviously he was the by far the best player in that draft
00:21:09
>> because international scouting wasn't then fully developed. Is that is that
00:21:13
what was going on? >> Yeah. And you know, we we had some international scouting, but we did not
00:21:17
have good international data uh at that. >> Okay. >> Yeah. >> Okay. So, Ben, you said this thing. You
00:21:23
said the odds have been flattened at the top of the lottery, and that was by design. And the design was to decrease
00:21:29
the incentive to have the top pick. But you're saying ironically in a deep draft, it gives exactly the wrong
00:21:36
incentives because if you can get into the top three or four, then you're have
00:21:41
e even chances at relatively even chances at a at one of the great players. Is that is that the
00:21:47
>> more So it means that if you're you know you know if you finish with the fifth,
00:21:51
sixth, seventh record, you now your odds of getting that first pick are higher than they used to be. So, you just want
00:21:57
to get down into that range. You don't have to get down to the bottom, but you
00:22:00
want to get down that range to have a reasonable shot at that pick. >> Um,
00:22:04
>> well, so these guys in the NBA front office are having a tough time, you
00:22:08
know, tweaking these numbers exactly uh in the way that would give the incentives to play throughout the
00:22:13
season. What do you think if if they they've made their best effort to get those the slope of those odds right,
00:22:19
it's still not working. What other levers can they pull? Well, I mean the so the there are there's one very clear
00:22:26
and then there's a second even more radical way to do this. The the the obvious way to do this to eliminate this
00:22:32
the the problem the incentive problem here is to just eliminate the draft. Um and you you have to you let all the
00:22:39
players coming in are they're all free agents and you go sign teams that want
00:22:42
them can go sign them. Uh and you eliminate the rookie scale contracts so players you know so teams are allowed to
00:22:49
differentiate their offers. You'd have to have a hard cap if you did that, right?
00:22:53
>> Well, you you you can have a hard cap and not um and not have a rookie scale.
00:22:58
Like that's fine. There's no reason those two things don't have to go
00:23:00
together, >> right? >> Um and so you can still have the hard cap and uh it, you know, unlikely that
00:23:07
the owners would get rid of that, but the the rookie scale um which you know would eliminate the use the value of the
00:23:13
going to the free agency. Once you have you eliminate the rookie scale, then teens can go and say, "All right, yeah,
00:23:18
I'm Utah. Uh, you may not want to come here, but I'm going to offer you a
00:23:22
million dollars more than LA is going to." So now I there there's way to
00:23:27
differentiate my offer from other teams. Uh, you know, or, you know, I'm going to
00:23:31
book, you know, invest in you to be the, you know, the franchise, the center of my franchise, whereas you if you go to
00:23:37
OKC, yeah, you're gonna you're going to be on the team, but, you know, you're
00:23:39
not going to be uh, you know, getting a lot of playing time right away. So, >> but Ben, in general that it is true
00:23:46
though that all the regional effects we were talking about before would would just be exacerbated, would they not?
00:23:52
Given that it's now now they have to compete also for the rookies in the same
00:23:56
way that they're disadvantaged when competing for free agents >> to to some extent except that uh one
00:24:01
there are only so much so many uh uh of these guys that teams are going to invest significantly in. So, if LA
00:24:08
decides that they want player X uh and and player Y, then that's it. Like, they're going to go after those guys,
00:24:14
they're not going to bid on every single player that in in the draft. Um they
00:24:19
can't. And or if you take a team that like, you know, if Boston is really good, uh then they're not going to spend
00:24:26
a lot on rookies, no matter how much people want to go there. They're just not going to because it's not worth it
00:24:30
to them. They need to save their cap for their existing star players. >> Um
00:24:35
>> that makes that that makes sense. But also you get the you get the intertemporal um effects as well. I
00:24:40
think I think because you're saying some drafts are much deeper than others. So
00:24:44
this upcoming draft anticipated being deep if in this in the world you're describing in the current world because
00:24:49
of the rookie cap the the slots are kind of fixed. The same amount is going to be
00:24:54
spent on the top 10 players regardless of whether they're great or whether they're lousy.
00:24:57
>> Exactly. >> In the world you're describing some years could be a lot more rookie pay
00:25:02
essentially than others according to what that's interesting. That alone sounds like a big inefficiency though.
00:25:07
>> So, but let me ask you a question. Let's imagine you're the you're the NBA.
00:25:10
>> Yeah. >> And Ben Alamar works for the NBA and suggests eliminating the draft.
00:25:16
>> Yeah. >> Is there any way you could kind of simulate or have a prediction of what
00:25:23
might happen if you got rid of the draft given it's never happened before? And so
00:25:28
I'm just wondering like could all hell break loose? Like how would you even think about evaluating a concept like
00:25:34
that? Yeah, I so I mean I think that what would happen and this is you know anytime there's a significant rule
00:25:39
change or you know this the the the uh the cap rules change significantly for whatever reason the smart teams figure
00:25:46
it out first. They figure out their strategy they they have a clear strategy that they execute around whatever the
00:25:52
new rules are. Um and the teams that aren't as as sophisticated or you know as as smart in the front office you know
00:25:58
they sort of like assume it's going to work exactly like free agency works now
00:26:02
for example. Um and and you know whatever the dynamics are they might there there's going to be some
00:26:07
interesting dynamics. We don't really know what they are. The closest thing we
00:26:11
have to this would be you know NIL and college basketball right now. Um where you know we are not seeing you know all
00:26:17
the players flocking to you know the biggest cities. They're flocking to the
00:26:20
teams that are going to pay them. Um there there's there's no cap there. you
00:26:24
know, all so it's it it's it's a little bit different obviously, but um you
00:26:29
know, team reputation, strength of the organization, >> all that opportunity for the player, all
00:26:34
those things are going to matter in that world. Um whereas, you know, right now in the draft, they don't because the
00:26:39
player gets no choice. >> By the way, guys, quick quick aside, uh in the WNBA, they've got this craziness
00:26:46
that's about to happen about something that's never happened before. They're
00:26:49
negotiating a new CBA right now. And for years, the agents have known that they're going to have a new CBA this
00:26:55
year. So, they've expired a lot of contracts now. >> And so, basically, almost every player
00:27:01
in the league is a free agent. As soon as they close the CBA, it's just going
00:27:05
to be a free-for-all. It's a complete free-for-all. And just as Ben says, the
00:27:08
advantage goes to the sharper teams, the teams that have been thinking about this
00:27:11
ahead of time. No one knows exactly what's going to happen, but can you prepare yourself for all the
00:27:16
contingencies so that you have an advantage in this new world? >> Ben, can I just have a counterargument
00:27:20
to yours? So, let's suppose I'm a player where I certainly care about money. I
00:27:23
mean, no doubt players care about money deeply, >> but they also care about winning.
00:27:27
>> So, if there's no draft and I'm I don't even know is one of the three players
00:27:31
one of the boozers. I don't even know. I know they is is are they one of the
00:27:34
players that you're talking about that's coming out? >> No, he's they've fallen out of the top
00:27:38
three. They're they're they're part of the high quality of this draft, but
00:27:41
they're not top they're not top. >> So, let's just say one of the top three
00:27:44
players says, you know, I care about money, but I do really care about winning. You know what? I'm gonna go to
00:27:49
OKC, San Antonio, those are my teams that I'm gonna go Denver. That's it.
00:27:55
Boston, maybe Detroit, New York. Six. Those are the six. Other 26 teams don't
00:27:59
even bother contacting me. Why wouldn't the rich get richer under this scenario
00:28:04
where you I'll just call it it's a multi- objective function where there's
00:28:09
some weight on winning and that's going to force possibly the over team the other teams to overpay
00:28:16
>> in uncertainty which could actually lead to their demise. >> Yeah. So I I mean so there there a
00:28:22
couple things there. One you've just named OKC and San Antonio is getting you
00:28:26
know the rich getting richer which is true but they're also pretty small market teams. So the small market big
00:28:32
market thing you know we it seems like that that's counter. The other sort of
00:28:36
part of that is like yeah if you're not if if you know winning is if if we assume that consistent winning is a
00:28:44
product of you know a strong smart organization um then that gives you know smart
00:28:52
organizations an advantage that players will want to go to those kinds of organization. Of
00:28:59
course, if you're if you're not a smart good organization that can't figure out
00:29:03
how to win, you're going to be at a disadvantage. And the only way to to to do it is you had to spend more. That's
00:29:09
the way to do it. Now, I while money matters, winning matters, opportunity matters also for these guys because
00:29:16
particularly the guys who are haven't, you know, proven may not have be off the
00:29:21
huge rookie uh offers. um they're going to have to prove themselves. And so if
00:29:25
they don't have opportunity to play going to OKC, Boston, uh you know, Denver, like if they're not going to
00:29:31
have an opportunity to play there, then they're not going to have an opportunity
00:29:33
to get their next contract, you know, right? And brilliant. >> So lots of dimensions going to come into
00:29:39
the position. >> Let me AI AI has a question. I have a followup on this topic as well. Audi, so
00:29:45
in the draft um you you're talking about uh or replacing it or something similar.
00:29:52
How much does the quality drop from the top three, top four and and that must make an enormous difference on this kind
00:30:00
of calculation? So I mean this is something that that uh we we've always talked about a moneyball. Um, and
00:30:08
basketball is clearly the most rapidly decaying function among all sports. And I'm not sure it's it actually is because
00:30:16
they really get that much worse or just because a team only has five players and
00:30:21
there's only one ball. And >> by the way, a do you mean a priori a posterior?
00:30:26
>> Uh, I mean, okay. I mean a priori. I mean basically I've just it's a great
00:30:33
question but I'm I'm essentially asking can you really build a team >> without having regular access to the top
00:30:41
X where X is what three five nine 20 um generally my understanding about the draft is after the first round is just
00:30:49
it's just you're just nothing there maybe you get a you know a generation
00:30:53
you get a player of any value every now and then but >> can you really build it without access
00:30:58
to the real Um, I mean, you you can build a team without access to the top, but you're
00:31:04
going >> winning team, right? >> You can build a team without access to
00:31:07
the very top of the draft, provided that you have other avenues to acquire, you know, high level talent. um you know,
00:31:16
you can't uh if if you don't if you can't bring players in through free
00:31:20
agency or trade that are of that caliber, the most likely place to get those kinds of players is at the very
00:31:27
top of the draft. They're not they don't all exist there. You can get them like
00:31:31
you go was, you know, and he's the exception that proves the rule on some level. I mean, you know, these these
00:31:38
players do exist and they happen, but they tend to be in the at least in the top 10 of the draft, uh, if not top
00:31:46
five. Uh, and so that doesn't it's really hard to compete without a couple of guys that are that
00:31:52
that good. >> Guys, let me real quick observation and then we need to change to a final
00:31:59
question for Ben before we go, but I just want to observe a specific thing about what Ben said before. If you did
00:32:04
go to the system, the the advantage to the sharp clubs essentially, if you think about it right now, the challenge
00:32:10
with a draft primarily is to order the players. And if you go to free agency model, the challenge is
00:32:18
to value every player. And the second task is much harder than the first task. And so that the advantage you if you're
00:32:26
a sharp club, you want a more complicated, more challenging task because you can separate more. And so
00:32:31
you might claim that that would be an advantage to the Sharp Club. Um Ben, I I saw you recently and you mentioned that
00:32:38
you've taken up a a cause. You're you're evangelical about a new cause. Yes.
00:32:44
>> That I think we'd probably be sympathetic to. So tell tell us about
00:32:46
this cause. Now Audi, you this is going to tickle you, I think. >> Okay. It it is my very strong belief that uh
00:32:54
that Dean Oliver belongs in the Basketball Hall of Fame. Uh I think I think the case is a slam dunk no
00:33:01
problem. Uh his qualifications and background, his impact on the game are so deep and obvious that um he he
00:33:10
absolutely uh belongs in in the Basketball Hall of Fame. Um Dean obviously wrote basketball on paper.
00:33:16
>> So hold on, Ben. Hold on. First tell us by what avenue do you get him in here?
00:33:20
We think about Will Chamberlain and Magic Johnson and Dean Oliver. Come on. >> Yeah.
00:33:24
>> Yeah. So, Dino, so there's a there's a a category in the Hall of Fame of
00:33:28
contributors, people who have had an positive impact on the game. Um, you know, and so those are the the that's
00:33:34
the avenue the the the lane in which Dean gets added to the Hall of Fame. >> And who are some examples from that?
00:33:40
>> I mean, there there there are variety of guys, but uh Dell Harris, Rod Thorne,
00:33:46
Rick Weltz, Doug Collins. Um, so these are and so you can get in multiple times. So, you know, uh, Doug Collins
00:33:53
can get in as a coach but also as a a contributor through his media work or, you know, and Del Harris has been a real
00:33:59
ambassador for the game overseas, for example. Um, uh, Rick Weltz has been really innovative on the business side
00:34:06
of of basketball and really contributes to the game. Um, Dean obviously has contributed to how teams think about
00:34:13
basketball and really changed that significantly. Um, so that that's why as a contributor, he really belongs in the
00:34:20
Hall of Fame. So, so Ben, tell us. So, you're you're we're analysts and so we
00:34:24
think Dean Dean speaks our language and so yes, he helps us understand it. He gives us a lens to understand the game.
00:34:30
You're from that same world. You have a PhD in economics. What's the evidence
00:34:34
that you would muster for his having changed the way actual basketball people who run teams and play the game think
00:34:42
differently because of Dean? >> Um the the whole concept of of efficiency and the four factors of
00:34:49
basketball uh are have permeated throughout the league. So Dean brought these concepts in basketball on paper
00:34:55
which is uh to me the the starting point of modern basketball analytics uh when he published that book. Um and so that
00:35:03
that I mean you find that book on on most you know in most front office on a bookshelf somewhere uh at least one copy
00:35:11
in every team building somewhere um and but so if you take the the four factors for example which is sort of a you know
00:35:19
clear way to think about a team uh you have their shooting efficiency, rebound efficiency, turnovers and ability to get
00:35:26
to the line. These are the four factors uh that that we can measure and you know
00:35:30
we know that we take those how a team performs on those things on offense and defense
00:35:35
90 plus percent correlation to to their point differential. >> Let me just point out because we I do
00:35:41
this with my my my senior uh students. I teach a sports analytics uh capstone and
00:35:47
it's uh remarkable because it works almost as well basically as well as point differential
00:35:54
>> just to show you. So, in other words, you're basically taking a bunch of
00:35:56
peripherals and you're using them to to predict winning percentage and it does
00:36:01
it as well as the score differential. What does this well, Audi? Not quite. It can't be exactly as well, but really
00:36:09
incredibly close. I mean, in fairness, shooting efficiency is one of them. And of course, that that's that's the
00:36:15
biggest factor. And it's not like and that doesn't that has a lot to do with
00:36:18
your score differential. >> So, real quickly to remind us what shooting efficiency is. And you
00:36:24
mentioned efficiency right off the top and if you want to give Dean credit for one thing it's probably this notion of
00:36:28
efficiency but to remind us what it is. >> So efficiency overall is points per
00:36:32
possession. That's how we think about and that that is a a a paradigm shift in
00:36:37
in basketball significantly. And shooting efficiency typically we think about effective field goal percentage
00:36:43
which says all right we're going to give you more credit for making a three-point
00:36:47
shot than a two-point shot because it's harder to make and it's worth more. Um
00:36:53
that's the the effective field goal percentage basically, you know, weights twos and threes more reasonably uh and
00:36:59
gives us a better sense of how well a team is shooting given these two different kinds of shots.
00:37:05
>> Well, Ben, by the way, if it makes you feel any better, I asked the Oracle chat
00:37:08
GPT what his odds are of making the Hall of Fame >> and it said actually it's pretty
00:37:15
moderate. It gave it 40% chance. Now, >> I could ask it on what basis and all
00:37:20
kinds of things, but it said he's an extraordinarily strong contributor, but
00:37:25
in a what people consider a niche area, but either way, >> but but your your Oracle doesn't know
00:37:30
that Ben Alamar is on the job. It hasn't picked up yet on the fact that there's
00:37:34
an evangelist out there. Who do you have to convince, Ben? And how and how long a
00:37:37
campaign are you geared up to wage? >> Yeah, so I'm I'm going to keep fighting
00:37:41
this one until uh I get uh get some traction. I fir I first rolled this out at Sloan a couple of years ago in a
00:37:47
history of basketball analytics panel uh that Dean was on, but I asked Mike Zaren, you know, when we were going to
00:37:52
get Dean in the Hall of Fame. Um and and Xarin uh acknowledged that that he belongs there. Uh
00:37:58
>> Xarn front office with the Boston Celtics. >> Absolutely. Yes. >> And call the Bill James of basketball.
00:38:06
So >> is Bill Bill James is not in the in the the baseball hall of fame? There's no
00:38:10
avenue for that, right? >> Yeah. I that that's you know >> if this avenue existed in baseball
00:38:16
>> I don't know how you do it but I don't >> he could be I mean Bill James would he
00:38:20
be cons I guess he wouldn't be considered a writer at all because there was obviously a writer's
00:38:24
>> I mean he's he he was a writer obviously >> he should be considered a writer and he
00:38:28
should be in the Hall of Fame. Absolutely why like they've got if there's not a reason for Bill for Bill
00:38:32
James be in the baseball hall of fame they got to figure that out because that's that doesn't seem right. By the
00:38:37
way, just so you know, um I asked Jot GPT if Ben Alamar really pushes for it. You added 10 to 15%.
00:38:45
>> I if I'm Ben, I'm not >> here's a better question, Eric. We're
00:38:49
out of control. >> I'm going to ask it pushes for it. Is that >> That's what I'm about to do. A I'm not
00:38:56
going to use me. I'm gonna use me. If Eric Bradloan says, "All right, now
00:38:59
we're down to 25." >> Yeah, we're losing. We're losing points.
00:39:02
>> But Ben, you're worth 10 out of it. Hey, hey, I want >> value ad.
00:39:06
>> We are getting the uh baseball hall of fame, the head of the baseball hall of
00:39:12
fame on the show in the next month and we're going to start the Bill James we're going to start the Bill James
00:39:18
campaign. >> There you go. Somebody needs to lead that charge. That's the I I I can't take
00:39:22
that one on, but I I'll uh >> actually, let me just say, Kade, I will ask um I will we're having Josh Trowick,
00:39:29
who's the president of the National Baseball Hall of Fame, on next week, Ben, and I'm going to ask him, do you
00:39:34
ever see a path where someone that's trains the game of baseball through analytics? Like, you could argue, why
00:39:40
not Billy Bean at some level, too? I mean, why doesn't he get why isn't he in
00:39:44
the Baseball Hall of Fame? >> What are you talking about? How about Theo Epstein?
00:39:48
>> Yeah. Well, that's another name. >> Championships, right? Theo will get
00:39:52
there, right? So, I mean, I to me I I mean, those guys like there's clear ways
00:39:58
for them to do it and and the the impact that they had are sort of advancing, you
00:40:02
know, using these things, you know, think about the front office in a new and interesting way with the tools. But,
00:40:08
you know, Bill James created the tools like Dean Oliver changed the way we, you know, make decisions in basketball. Um,
00:40:14
and and that's >> it's a great cause. It's a great cause. Keep beating that drum. We'll we'll
00:40:19
start we'll start going along with you. Ben, we've kept you longer than we meant
00:40:22
to. Thank you very much. Always a delight to talk to you. >> My pleasure. Thanks, guys.
00:40:27
>> Ben Alamar, longtime sports analytics. He has got a book on sports analytics.
00:40:32
He is teaching a program on teach on sports analytics at the University of Texas at the moment. He will be at the
00:40:39
Sloan Sports Analytics Conference hosting a panel in just about a month's time. All right, guys. That has been the
00:40:45
first half of Orton Moneyball. Come back and join us after the break. Welcome back.
00:40:51
Welcome back to Wharton Moneyball. Welcome to the second half or you might say the last quarter. We went long that
00:40:57
first first half with Ben Alamar. Delightful conversation. Could have gone on for a while. Ben's always a fun
00:41:04
conversation. We're we got to go about 15 minutes. Audi has to get the classroom. So, let's do a quick What
00:41:10
caught your eye. There are obviously interesting things floating around. I think Audi might want to jump in here
00:41:15
first. gentlemen around the world. >> I will jump in because because it's
00:41:19
something that I've studied in general though not in specifics. It's the it's
00:41:22
the judging fiasco in skating at the Olympics. >> Okay. >> And uh so I won't fill in all the
00:41:29
details but suffice it to say the French judge um gave a very high score to the French competitors and importantly gave
00:41:37
an extremely low score to the American competitors and it essentially completely reversed the order of the the
00:41:44
competition. Which competition was this? >> And of course, this was the this is the
00:41:48
doubles the mixed doubles skating competition. Um I forget exactly which version. And as you know, Shane's not
00:41:55
here to scream and yell, but he'll he'll tell you that real sports aren't
00:41:59
subjective, right? Um but of course, skating is a real sport and it's has judges and they have a format and
00:42:06
typically you have a set of judges. each each make a score and then there's a
00:42:10
then there's a a compet some sort of summary of the usually >> don't they trimmed uh
00:42:16
>> No, I know they do that in diving. Um I'm not sure they do it in in skating.
00:42:20
>> I thought they did it in skating. I thought they >> I think I think this I I I think this
00:42:24
might be a niche skating competition, not the main skating competition. And I feel like they've really ironed out the
00:42:30
main skating competition in a way that I mean I can tell you that uh >> it's like ice dancing or something
00:42:36
instead of the actual Well, I think the ice danders would disagree whether or not it's niche or not. May not be the
00:42:42
historically the most famous of the >> But Audi, there were full-on like communist block almost battles over this
00:42:50
back years ago. I mean, the whole the whole meme of the Russian judge comes from that, right?
00:42:55
>> Yeah. But this they ironed it out in some way. So, this must be >> Well, clearly they haven't because
00:42:59
there's a controversy over this and I haven't looked at the data. But one of
00:43:03
the things that I think as a statistician is interesting is the idea that you can essentially
00:43:08
um establish an enormous bias you um I don't know what you want to call it other than that
00:43:14
a if without having much data um if the delta is ex is exceptionally large relative to historical standards and I
00:43:23
believe that is what happened here. I see. Well, let me just say I did look. They do prune. They prune the highest
00:43:28
and the lowest. There's nine judges for both technical and whatever they call
00:43:32
the other one aesthetics or whatever. They do prune the highest and lowest. So, one judge, even a French judge on a
00:43:39
French team could not actually lift them, although it does remove margin of error if you want. It means, you know,
00:43:46
that means the second score is going to count in some way much more than otherwise. So I'm not
00:43:52
>> what would have been the max now becomes the second and that gets >> that's what I'm saying the distribution
00:43:57
and score is different than the max. >> You're saying the delta you're saying
00:44:00
between the top rated and lowest rating is that the delta you're talking about
00:44:03
what delta >> delta delta is between the top and the bottom compared to the the overall panel
00:44:09
average >> of the delta the panel average >> that's what I would call the delta. So
00:44:13
there's a, you know, one of the one of the exercises that I've done with
00:44:16
students is you can find a judge that only has two or three uh measurements where the he has a nationality match
00:44:25
with his compatriate during the competition. And you can still nevertheless demonstrate pretty clearly
00:44:31
that there is bias because he had he might have judged 40 or 50 skaters or divers or whatever competition you're
00:44:38
looking at and his two highest differentials or delta are for when he made the match and that just has
00:44:44
astoundingly small probability. >> Why don't they just knock out the judges
00:44:48
rating their own nation? Why is that? That would be an interesting Well, you can't that easily because the nations
00:44:56
with the largest competitors, United States or China, France, whatever it is, probably have more than one. And that
00:45:02
would would um >> Well, don't give them more than one. There's plenty of nations with figure
00:45:07
skating going on to have one per >> right from different >> Anyway, it's not done. At least not yet.
00:45:13
>> Interesting. >> All right, Eric, what you got? >> Well, I got a brief one and then I I
00:45:18
still want to emphasize again. I'm a big fan because I used to ski a lot. I'm a
00:45:23
big fan of skiing and I want to say it again. So now Michaela Shiffren, the most celebrated women's downhill skier
00:45:31
in the history of skiing. Matter of fact, of male or female, is now 0 and8 in her last eight Olympic races. And I
00:45:39
just want to be clear, I don't mean 0 and8 not winning the gold. I mean 0 and8
00:45:43
not even medlin. And so, look, I the good news is, let me say for her, she has two golds and a silver from the
00:45:52
first three races she ever did in the Olympics. That's great. That's an extraordinarily accomplished career, but
00:46:00
it's starting to get to the point where it is going to detract from her overall
00:46:06
accomplishments as a skier. It's just going to This is about to be She has one race left. It's
00:46:12
about to be the third straight Olympics where she will not win a medal. And >> so so in bet so the race is tomorrow's
00:46:19
Wednesday slalom I believe her best >> her best event. So what is your bet?
00:46:26
What probability do you give that she medals? >> I would say one/3. >> Okay. So it's distinctly below what her
00:46:39
world ranking would suggest. below what her world ranking is, but still high enough that reflects that that is her
00:46:46
best event and she's the best slalom women's slalom racer of all time. >> Uh, do you go you want you want above or
00:46:53
below 33%. Who's you want to be above >> winning? Is this winning or meddling
00:46:58
meddling? Mattling >> meddling. I want above it. >> I want to I want above it, too. So, you
00:47:02
and I neither one are going as far to the to the >> I'm not Well, you know, we are not
00:47:07
momentum guys. >> I'm a momentum person. I mean, let's just get a meme going here.
00:47:11
>> Yeah, let's go. But let me talk about the topic I really wanted to talk about.
00:47:14
>> So, no, no, hold on. We're real quickly on this point. Um, you guys are hardcore
00:47:19
statisticians. Do you believe people that practice like performance coaches are a thing? Do you think that's
00:47:26
a real Do you think they can Do you think they can substantively shift an athletes performance? A performance
00:47:31
coach, a mental coach? God, you know what? I have nothing substantive to say on that on that
00:47:39
statement. >> I I I don't know. >> Well done, Audi. Well done. That's good
00:47:43
awareness. >> Do I think a performance coach can help someone perform at their maximum
00:47:52
>> given also understanding there is interday variation and all of those things? Yes, I do. I I I do. I believe I
00:48:00
didn't say that they can make them necessarily perform better, but can they make someone perform near their maximum
00:48:07
maybe more consistently? Yes, >> that's that's yeah, that that's that's
00:48:11
really well defined. You you operationalized it better. >> Yes, I think we should we should find we
00:48:16
should find someone to come on and talk to us about that. You're hearing lots of
00:48:19
between what's happened with Shiffron and the figure skater Ilia, what's his
00:48:23
name? There's lots of talking about that. Eric, I've got to sneak mine in
00:48:27
here before Audi goes away. Okay. So, if you want to linger for a minute, I'll
00:48:29
get your second one. Let me let me put one in there that I that that I have for y'all in particular.
00:48:34
>> I want to hire y'all. Y'all y'all do speaking, y'all do statistical
00:48:38
consulting on occasion. I want to hire y'all. I run the Olympics and I'm
00:48:41
worried about the the I think it's been kind of willy-nilly uh when events when
00:48:46
when we get multiple versions of the same event and it's not clear to me why
00:48:51
we should have multiple versions or maybe we should have more, maybe we should have fewer. So I want to know
00:48:55
like is four the right number of downhill events to have. We have a downhill on one end, we have slalom on
00:49:02
the other, but now we have a giant slalom, we have a super G. So we have four variations and I don't know I run
00:49:07
the Olympics. Is four the right number? I mean is four too much? Should it be five? Okay, different var different
00:49:13
question. Speed skating and this is this goes back to something Audi's complained
00:49:17
about in the in the summer Olympics. How many different events there are with swimming. So speed skating, what is the
00:49:22
right number of links to have? Not forget short track, just traditional speed skating. There's 500, there's
00:49:29
thousand, there's 1500, 5,000, 10,000. And I I run the Olympics. I don't know.
00:49:34
Are these speed skaters giving us all these extra races because they want more medals? Is it does it make sense? Or
00:49:39
maybe we need even more. Maybe we need a 750 meter. I need y'all as statisticians
00:49:44
to tell me how I should think about this, >> bro. So I can start with one way to
00:49:49
think about it which which uh I know has is similar in track and it's certainly
00:49:53
true in swimming is that there's a huge difference between the training the body
00:49:58
type the skill of being distance to sprint and and the middle is also essentially its own it its own
00:50:06
centroidid now and the reason why that's fascinating is that basically you have
00:50:11
the elite in each of those three categories and they're almost always distinctly different people except for
00:50:17
an occasional superstar can dominate in two over you know two adjacent uh lengths. So in swimming you can have
00:50:26
someone who's good at the 400 and maybe the 800 >> Katy Leiddki >> Katy Leiddki is the only one who really
00:50:32
comes to mind. Um, for example, uh, even in >> Okay, so now I I understand the theory.
00:50:38
>> I've got a >> Okay, I understand that I I understand the >> I have a different I'm sorry. Let me
00:50:42
just jump in quickly about different measure. A different measure would be let's imagine you could measure ability
00:50:49
at different >> uh lengths. And let's imagine I'm just trying to operationalize this. Let's
00:50:56
imagine the correlation between theta j and theta j + one where j indicates the length of the race was extraordinarily
00:51:04
high. So that in some sense you might as well call it one winner because the ability is so correlated between the to
00:51:13
me that would suggest a redundancy in races. >> So your what you did with your theta was
00:51:19
the underlying ability it's theta j is that right? So it's all the different
00:51:23
thetas for the the >> it's your theta was the underlying ability. You're going to estimate the
00:51:30
underlying ability, but it's theta i. You're going to do underlying thetas for
00:51:33
all the different skiers. And then you're going to do it for each of the different races. And you're going to ask
00:51:38
how correlated the theta i's are across J's. So for for each skier, what their theta in an event, how is it
00:51:47
correlate with their theta in another event? Do I have that right? That's that's a very clear
00:51:52
That's what that's what my suggestion was. >> Okay. Okay. So So take take downhill. So
00:52:00
um Audi, are you happy with that? And by the way, what what what would your criteria be? How how how would we know
00:52:06
what's high or what's low? Like take the four downhill events and so how are we
00:52:10
going to judge whether that's do we really need Super G or is it like you know the giant slalom is enough? That's
00:52:17
the middle race. Super G was I think the last last comer. I'm just statistically
00:52:22
or >> so now you want me to come up with a cutoff score you're saying?
00:52:26
>> Well, I just want I'm asking Audi, I like the way you did that, but I want to
00:52:29
know Audi, are you satisfied with that? I thought there might be a clust is clustering any different really than
00:52:35
what Eric just suggested. >> I don't think so. I mean, it's complicated issue because there's also
00:52:41
the marketing side of this. people want to see more events because it's just
00:52:44
more fun, more opportunity >> and that really giant giantly competes. I mean I mean speed skating you might
00:52:48
say is over is overly overly divided too many because think about Eric Heiden I think he won seven gold medals in um
00:52:56
>> but he was making that many >> that was 46 years ago and one of the
00:53:00
things they have said about this Wisconsin kid is that he was the first only the second ever to win both the 500
00:53:06
and the 1000. So that's Eric's correlation not being very high the 500 and 1000. So apparently that's a
00:53:13
distinction between the short and the middle that's substantive. Okay, I'll
00:53:16
let that go. I want to get your quick thoughts and we should probably cut odd out while
00:53:22
>> Eric is going to give us another what he really want to talk about. >> Stay and I'll talk to you next week.
00:53:26
Have fun in class. >> So I just had one more. So this number this statistic has been on my mind for
00:53:32
weeks now because it's been true. So I'll have you guess for a second. So the
00:53:36
Lakers by the way this year it's in the NBA of course have a very good record.
00:53:39
They're 33 and 21. I think we'd agree that's a good record, right? >> Good.
00:53:44
>> It's good. I didn't say it's great. I said it's good. It's a good record.
00:53:47
They're actually winning I mean 54 games. Uh they're winning well more than
00:53:52
60% of their games. So they're playing well. >> Do you know what their point
00:53:55
differential is, Kate? >> I'm guessing not very much. >> It's zero.
00:53:59
>> This Okay. >> So I compare that to teams around them like the Nuggets,
00:54:07
the Rockets, etc. their differential is like plus five. So over 50s something games those
00:54:15
teams have more than a 300 like a 300 point differential and the Lakers have zero.
00:54:20
>> So you're saying the average the average differential is.5. >> Yeah. The average per game. The average
00:54:24
Yeah. per game. >> Well, this is this is Pythagorean. This is the Pythagorean.
00:54:28
>> That's what I'm saying. This to me seems like a a significant deviation from
00:54:32
Pythagorean where literally they're at zero, but somehow whether you could argue they're I mean there's two things
00:54:40
that could cause us, right? They're winning a lot of close games and LeBron James is going to help you do that
00:54:45
>> or they're getting blown out a lot in games, which is when they lose, they
00:54:51
lose real bad. And by the way, that's cuz Luca hasn't played a lot of games.
00:54:54
LeBron's injured. But all I'm commenting on is this is something actually next
00:54:59
time I speak to you know Ben Alamar or you know Dean Oliver or somebody else on I'm going to I I'm going to ask him like
00:55:07
how much does because again OKC let me just say by the way despite them having played 500 ball over the last 20 games.
00:55:14
If we look over the almost 60 games of the season so far the season ended right now they would break the record their
00:55:20
own record of last year for the greatest point differential in NBA history. per game. So at some level that has to
00:55:28
say something. Just in the same way to me the Lakers being at zero means >> they're really 500ish but they just
00:55:36
happen to be 33 and 21 unless someone tells me otherwise. >> One one version of this question would
00:55:43
be is the Pythagorean theorem which comes out of Bill James. We talked about Bill in the first half the show is a
00:55:49
Bill James thing but then it gets applied across all sports. It's basically just asking
00:55:53
>> to what extent it's saying in a big sample your point differential is going
00:55:57
to relate to your win record. And so we we kind of know whether someone's underperforming or under overperforming
00:56:04
their um their um point differential. Um here's one version of the question,
00:56:09
Eric. Might it mean less in basketball than other sports? I guess that's you've
00:56:14
already hypothesized on why the Lakers might have a worse differential than their
00:56:19
record would suggest because they've got a couple of players that are key and
00:56:22
those players aren't playing as regularly and so they have these disproportionate negative point
00:56:27
differentials when those guys are out. Might that be just kind of a hallmark of basketball these the load issues and the
00:56:34
impact of a single player and stars? might all that suggest that Pythagorean would be less predicted than
00:56:40
>> that's by the way that's an answerable uh empirical question and it's a good
00:56:44
one. It's also one related to the topic we talked to Ben Alamar about in the
00:56:48
first half. You you have a league with tanking and a team with resting players. You even mentioned resting players and
00:56:56
you're you're going to get situations like here's here's you know here's an
00:57:00
argument. What we're observing for the Lakers is a mixture distribution between
00:57:04
the 80% of games they're trying, the 20% of games they're not. That leads to a
00:57:11
breakage of the Pythagorean formula and it leads to a point differential of zero, but a massively winning record.
00:57:19
>> Okay? And that that to me suggests that baseball I mean basketball needs these mixture distributions more
00:57:27
than other sports do. Like you're not gonna get discreetly different distributions of team performance in the
00:57:34
NFL like you will in basketball and you're certainly not going to get it in baseball either, right? So mixed
00:57:40
distribut. >> Okay. So if that's the case then for sure it will be less meaningful a
00:57:50
Pythagorean. It's almost like you need you need to have a mixture of your Pythagoreans.
00:57:54
>> I order I like it actually. you in some sense you could fit you could imagine
00:57:59
fitting a latent class or a mixture model unless you want to say for basketball you can kind of tell based on
00:58:04
who they play which are the real games and which are the tank games you then fit a model separately or fit the
00:58:11
formula separately to those two and then see what would happen. I think that would be a that would be something
00:58:16
interesting. >> Well, it calls to mind uh when Nate Silver was publishing at 538, they had a
00:58:22
basketball model. He probably still publishes a basketball model, but they had um a basketball power ranking and
00:58:28
they had two versions of it, Eric. They had regular season and playoff. And they
00:58:32
were re that's that's essentially the mixture distribution we're talking
00:58:35
about. They're recognizing that there are there's a full strength team in the
00:58:39
NBA and then there's some non full strength team and you need to understand them and model them separately.
00:58:45
>> And just I know we're going to run out of time, but just one last thing to give
00:58:48
a shout out for your Texas boy Scotty Sheffller. I mean this I don't care that
00:58:54
he didn't win. It was great theater. This is the greatest in golf since Tiger
00:58:59
Woods in like 2001. He's now got 18 top 10s in a row. I mean, come on. >> I mean, unbelievable. And you know, um,
00:59:10
and it's amazing how great a golfer this man is. I just I and I was very happy to
00:59:16
see Colin Maro win his first tournament in almost three years. you know, he won >> by age 24 and hasn't even won since then
00:59:23
essentially. I was thrilled to see more win. But watching Sheffller blow three putts 5 feet or less and still shot a 63
00:59:31
by the way, including making three Eagles. I don't think in my whole life of watching golf I've ever seen a golfer
00:59:37
make three Eagles before in one single round of golf. I'm sure it's happened,
00:59:41
but I'm just commenting that >> I think we we should appreciate Scotty
00:59:46
Sheffller for who he is. And I'm gonna tell you, he's he's not Tiger Woods yet.
00:59:51
Not even close. Let him do it for another five years. But I have no reason to believe he won't do it for another
00:59:56
five years. And look, I don't know if he's getting to 15 majors. That's not
01:00:00
fair. But he could easily get into the Gary Player, uh Tom Watson, you know, that that league of 8 to 10 majors. And
01:00:09
I'm going to tell you something, this is a top 10 golfer of all time. He's great.
01:00:15
that I I don't I don't disagree with any of that, Eric. It reminds me it makes me
01:00:19
wonder about what other comparisons we can come up with. Um like we let's let's
01:00:24
let's move beyond Tiger because Tiger is well within our adult memory and we
01:00:29
lived that we lived it. We understand what that's like. There were a handful
01:00:33
of great golfers in our lifetime, but we probably weren't paying enough attention
01:00:36
and the metrics weren't there yet. And then there are a number of them before
01:00:39
our lifetime. If what's if you go back and ask what's the best five-year run of
01:00:44
golf in PGA history, like nine of the top 10 are Tiger. So, let's not do it that way. Let's do it. Let's do it at a
01:00:51
golfer level and ask peak five years of the best 10 or 12 golfers in the history
01:00:58
of golf and ask what does Scotty need to do or where to pick three years. Pick three years like Nicholas at his best,
01:01:07
Watson at his best. How is Sheffer comparing to that? And let's let's let's
01:01:12
stack up the historical golfers again. Tiger's going to be number one. And if
01:01:15
we let him have multiple observations, he'd be the top 10. Yeah. But let's not
01:01:19
do that. And let's just ask because we're not I want the next comparison.
01:01:22
>> Yeah. I I'm going to give you that. But just one last thing. I think the thing
01:01:24
that's amazing about Schoffler, besides he has 20 wins already in his career, he
01:01:28
had his first one only four years ago. So let's just be clear. He's had a
01:01:33
four-year run. That's incredible. I think it's fair to compare him to the
01:01:37
three-ear stretch. Let's talk about recent golf. The four the 3 to four year
01:01:41
stretch that Rory Maroy had when he piled up all those majors but hadn't still yet won the Masters. Remember he
01:01:48
went 11 years without winning a major between ages 21 and 24. Rory Maroy cleaned up. Let's even say Jordan.
01:01:55
>> How many wins How many wins did he have? >> That I don't know. But I'm guessing
01:01:59
didn't come close to it's not close to 20. He might have had as many majors but
01:02:04
I'm not even sure of that. >> No, no, we want we want the other as well. Well, this is a question like
01:02:07
what's the criteria? >> Yeah, but you ask who are the other stretches? I would compare it to Rory
01:02:12
Mroy during that stretch. Jordan speed that stretch. Those would be the >> he's going to be speed. No, no, it's
01:02:17
going to be he's going to beat those guys. So, but it's it's not Tiger, but
01:02:20
it's I want to know how he compares to Nicholas's best to Palmer's best and
01:02:25
going further back like Hogan's best like what? >> I have that for you and I will announce
01:02:32
it. Not next week. Maybe I will in the open segment, but I'm so excited about
01:02:35
the Baseball Hall of Fame guy next week. But I will I will produce that analysis
01:02:40
on request. >> Let's have a moment on the criteria and then we'll let it go.
01:02:45
>> Um we've already decided it can't just be majors and it can't just be wins. It
01:02:49
could be wins. It could be number of top tens, but we like wins and that's what
01:02:53
they care about is wins. Um I don't know. We could have we could have the actual, you know, um uh scoring average
01:03:02
below the PGA average. >> We're going to need multiple dimensions. >> Yeah.
01:03:08
>> Um wins would be the easiest like the the probability of winning like the like
01:03:13
the six month probability of winning. Um something like that. But anyway, fun conversation. We need to push beyond the
01:03:23
tiger comparison. Um, I just want to know how and we have to we can't do it from memory is one of the things that my
01:03:30
sense is. So, let's let's ask about Nicholas and Watson in their prime and a
01:03:34
few others like that. Okay, why don't we wrap it there? That has been a full more
01:03:40
than a full hour of sports analytics here on Wharton Moneyball for the whole team. Eric Bradloow who's been in here
01:03:45
for the whole run. Audi Winer was here for almost all of it. Shane Jensen in Absentia, our friend and associate
01:03:52
producer Dion Simkins who makes things happen around here, Marissa Raina, our producer, and Deep Patel, the boss lady.
01:03:59
Many thanks to you guys for listening as well. Come back and join us next time between now and then. Enjoy your sports.