Category Archives: Gambling

A Reason Bill Simmons is Bad At Gambling

For those unaware, Bill Simmons, aka the Sports Guy, is the editor-in-chief of Grantland, ESPN’s more literary (or perhaps intelligent, if you prefer) offshoot. He’s hired a lot of really excellent  writers (Jonah Keri and Zach Lowe, just to name two), but he continues to publish long, rambling football columns with limited empirical support. I find this somewhat frustrating given that the chief Grantland NFL writer, Bill Barnwell, is probably the most prominent data-oriented football writer around, but you take the good with the bad.

Simmons writes a column with NFL picks each week during the season, and has a pretty so-so track record for picking against the spread, as detailed in the first footnote to this article here. Simmons has also written a number of lengthy columns attempting to construct a system for gambling on the playoffs, and hasn’t done too great in this regard either. I’ve been meaning to mine some of these for a post for a while now, and since he’s written two such posts this year already (wild card and divisional round), I figured the time was right to look at some of his assertions.

The one I keyed on was this one, from two weeks ago:

SUGGESTION NO. 6: “Before you pick a team, just make sure Marty Schottenheimer, Herm Edwards, Wade Phillips, Norv Turner, Andy Reid, Anyone Named Mike, Anyone Described As Andy Reid’s Pupil and Anyone With the Last Name Mora” Isn’t Coaching Them.

I made this tweak in 2010 and feel good about it — especially when the “Anyone Named Mike” rule miraculously covers the Always Shaky Mike McCarthy and Mike “You Know What?” McCoy (both involved this weekend!) as well as Mike Smith, Mike “The Sideline Karma Gods Put A Curse On Me” Tomlin, Mike Munchak and the recently fired Mike Shanahan. We’re also covered if Mike Shula, Mike Martz, Mike Mularkey, Mike Tice or Mike Sherman ever make comebacks. I’m not saying you bet against the Mikes — just be psychotically careful with them. As for Andy Reid … we’ll get to him in a second.

That was written before the playoffs—after Round 1, he said he thinks he might make it an ironclad rule (with “Reid’s name…[in] 18-point font,” no less).

Now, these coaches certainly have a reputation for performing poorly under pressure and making poor decisions regarding timeouts, challenges, etc., but do they actually perform worse against the spread? I set out to find this out, using the always-helpful pro-football-reference database of historical gambling lines to get historical ATS performance for each coach he mentions. (One caveat here: the data only list closing lines, so I can’t evaluate how the coaches did compared to opening spreads, nor how much the line moved, which could in theory be useful to evaluate these ideas as well.) The table below lists the results:

Playoff Performance Against the Spread by Select Coaches
Coach Win Loss Named By Simmons Notes
Childress 2 1 No Andy Reid Coaching Tree
Ditka 6 6 No Named Mike
Edwards 3 3 Yes
Frazier 0 1 No Andy Reid Coaching Tree
Holmgren 13 9 No Named Mike
John Harbaugh 9 4 No Andy Reid Coaching Tree
Martz 2 5 Yes Named Mike
McCarthy 6 4 Yes Named Mike
Mora Jr. 1 1 Yes
Mora Sr. 0 6 Yes
Phillips 1 5 Yes
Reid 11 8 Yes
Schotteinheimer 4 13 Yes
Shanahan 7 6 Yes Named Mike
Sherman 2 4 Yes Named Mike
Smith 1 4 Yes Named Mike
Tice 1 1 Yes Named Mike
Tomlin 5 3 Yes Named Mike
Turner 6 2 Yes

A few notes: first, I’ve omitted pushes from these numbers, as PFR only lists two (both for Mike Holmgren). Second, the Reid coaching tree includes the three NFL coaches who served as assistants under Reid who coached an NFL playoff game before this postseason. Whether or not you think of them as Reid’s pupils is subjective, but it seems to me that doing it any other way is going to either turn into circular reasoning or cherry-picking. Third, my list of coaches named Mike is all NFL coaches referred to as Mike by Wikipedia who coached at least one playoff game, with the exception of Mike Holovak, who coached in the AFL in the 1960s and who thus a) seems old enough not to be relevant to this heuristic and b) is old enough that there isn’t point spread data for his playoff game on PFR, anyhow.

So, obviously some of these guys have had some poor performances against the spread: standouts include Jim Mora, Sr. at 0-6 and Marty Schottenheimer at 4-13, though the latter isn’t actually statistically significantly different from a .500 winning percentage (p = 0.052). More surprising, given Simmons’s emphasis on him, is the fact that Reid is actually over .500 lifetime in the playoffs against the spread. (That’s the point estimate, anyway; it’s not statistically significantly better, however.) This seems to me to be something you would want to check before making it part of your gambling platform, but that disconnect probably explains both why I don’t gamble on football and why Simmons seems to be poor at it. (Not that his rule has necessarily done him wrong, but drawing big conclusions on limited or contradictory evidence seems like a good way to lose a lot of money.)

Are there any broader trends we can pick up? Looking at Simmons’s suggestion, I can think of a few different sets we might want to look at:

  1. Every coach he lists by name.
  2. Every coach he lists by name, plus the Reid coaching tree.
  3. Every coach he lists by name, plus the unnamed Mikes.
  4. Every coach he lists by name, plus the Reid coaching tree and the unnamed Mikes.

A table with those results is below.

Combined Against the Spread Results for Different Groups of Coaches Cited By Simmons
Set of Coaches Number of Coaches in Set Wins Losses Winning Percentage p-Value
Named 14 50 65 43.48 0.19
Named + Reid 17 61 71 46.21 0.43
Named + Mikes 16 69 80 46.31 0.41
All 19 80 86 48.19 0.70

As a refresher, the p-value is the probability that we would observe a result as or more extreme as the observed result if there were no true effect, i.e. the selected coaches are actually average against the spread. (Here’s the Wikipedia article.) Since none of these are significant even at the 0.1 level (which is generally the lowest barrier to treating a result as meaningful), we wouldn’t conclude that any of Simmons’s postulated sets are actually worse than average ATS in the playoffs. It is true that these groups have done worse than average, but the margins aren’t huge and the samples are small, so without a lot more evidence I’m inclined to think that there isn’t any effect here. These coaches might not have been very successful in the playoffs, but any effect seems to be built into the lines.

Did Simmons actually follow his own suggestion this postseason? Well, he picked against Reid, for Mike McCoy (first postseason game), and against Mike McCarthy in the wild card round, going 1-0-2, with the one win being in the game he went against his own rule. For the divisional round, he’s gone against Ron Rivera (first postseason game, in the Reid coaching tree) and against Mike McCoy, sticking with his metric. Both of those games are today, so as I type we don’t know the results, but whatever they are, I bet they have next to nothing to do with Rivera’s relationship to Reid or McCoy’s given name.


Tied Up in Knots

Apologies for the gap between posts–travel and whatnot. I’ll hopefully have some shiny new content in the future. A narrow-minded, two part post inspired by the Bears game against the Vikings today:

Part I: The line going into the game was pick ’em, meaning no favorite. This means that a tie (very much on the table) would have resulted in a push. Has a tie game ever resulted in a push before?

As it turns out, using Pro Football Reference’s search function, there have been 19 ties since the overtime rule was introduced in the NFL in 1974, and none of them were pick ’em. (Note: PFR only has lines going back to the mid-1970s, so for two games I had to find out if there was a favorite from a Google News archive search.) (EDIT: Based on some search issues I’ve had, PFR may not list any games as pick ’ems. However, all of the lines were at least 2.5 points, so if there’s a recording error it isn’t responsible for this.)

Part II has to do with ties, specifically consecutive ones. Since 1974, unsurprisingly, no team has tied consecutive games. Were the Vikings, who were 24 seconds 1:47 shy of a second tie, the closest?

Only two teams before the Vikes have even had a stretch of two overtime games with one tie, both in 1986. The Eagles won a game on a QB sneak at 8:07 of OT a week before their tie, in a game that seems very odd now–the Raiders fumbled at the Philly 15 and had it taken back to the Raiders’ 4, after which the Eagles had Randall Cunningham punch it in. Given that the coaches today chose to go with field goal tries of 45+ even before 4th down, it’s clear that risk calculations with respect to kicking have changed quite a bit.

As for the other team, the 49ers lost on a field goal less than four minutes into overtime the week before their 1986 tie. Thus, the Vikings seem to have come well closer to consecutive ties than any other team since the merger.

Finally, a crude estimate of the probability a team would tie two consecutive games in a row. (Caveats follow at the end of the piece.) Assuming everything is independent (though realistically it’s not), we figure a tie occurs roughly 0.207% of the time, or roughly 2 ties for every thousand games played. Once again assuming independence (i.e. that a team that has tied once is no more likely to tie than any other), we figure the probability of consecutive ties in any given pair of games to be 0.0004%, or 1 in 232,000. Given the current status of an 32 team league in which each team plays 16 games, there are 480 such pairs of games per year.

Ignoring the fact that a tie has to have two teams (not a huge deal given the small probabilities we’re talking about), we would figure there is about a 0.2% chance that a team in the NFL will have two consecutive ties in a given year, meaning that we’d expect 500 seasons in the current format to be played before we get a streak like that.

I’ll note (warning: dull stuff follows) that there are some probably silly assumptions that went into these calculations, some of which—the ones relating to independence—I’ve already mentioned. I imagine that baseline tie rate is probably wrong, and I imagine it’s high. I can think of two things that would make me underestimate the likelihood of a tie: one is the new rules, which by reducing the amount of sudden death increase the probability that teams tie. The other is that I’ve assumed there’s no heterogeneity across teams in tie rates, and that’s just silly—a team with a bad offense and good defense, i.e. one that plays low scoring games, is more likely to play close games and more likely to have a scoreless OT. Teams that play outside, given the greater difficulty of field goal kicking, probably have a similar effect. Some math using Jensen’s inequality tells us that the heterogeneity will probably increase the likelihood that one team will do it.

However, those two changes will have a much smaller impact, I expect, than that of increasing field goal conversion rates and a dramatic increase in both overall points scored and the amount of passing that occurs, which makes it easier for teams to get more possessions in one OT. Given the extreme rarity of the tie, I don’t know how to empirically verify these suppositions (though I’d love to see a good simulation of these effects, but I don’t know of anyone who has one for this specific a scenario), but I’ll put it this way: I wouldn’t put money down at 400-1 that a team would tie twice in a row in a given year. I don’t even think I’d do it at 1000-1, but I’d certainly think about it.

Lining Up Behind the 76ers

Going into the year, there was an honest discussion as to whether the 76ers would post the worst record in NBA history, topping their own record of 9-73 from 1972-73. (See here and here.) After 3 games…well, let’s just say that discussion’s been tabled. 3 wins in a month makes it hard to take a run at 8-74, especially when two of them are against the very best teams in the league. It’s going to take serious commitment to winning the Ender Game* for them to not scrape out another 6+ wins playing against the Raptors, Celtics, et al.

Still, I would say that the story is not that the Sixers aren’t historically terrible—it’s that they’re quite terrible and managed to beat three teams in four nights, two of which were expected to be very good.

Just how unexpected was this? There are obviously a few ways of looking at this. I could look to AccuScore, probably the most notable sports prediction engine out there, though I can’t find their early season NBA picks lying around anywhere. However, everything is more fun when there’s money involved, so we’re going to do this in terms of gambling. Specifically, if you bet the Sixers to beat the Heat straight up, then bet all your winnings on them against the Wizards, and did it again against the Bulls, how would you have done? Moreover, how often does a streak that’s this improbable occur?

Well, according to Odds Portal, if you’d started with $100 and kept reinvesting, you’d have a stake of $13,206.26, for a profit of $13,106.26. If you prefer a percentage return, since we picked $100 you can see this pretty easily—131x, or 13100%. This isn’t unheard of in sports gambling—someone made $375K on two $250 bets on the Cardinals in 2011—but it’s still quite impressive, especially given that these are single game bets rather than bets on a team winning the title.

How impressive is it? To answer that, I scraped NBA money line data from 2007-08 through 2011-12 from Sports Book Reviews, and while I can’t evaluate the exact accuracy of their lines, I figure it’s probably good enough. (They seem a little extreme, but I don’t bet enough basketball to say for sure.) I looked at every three game winning streak in that dataset, counting longer streaks multiple times, e.g. a five game winning streak is three overlapping three game streaks. (Playoffs and multiseason streaks are also included, though neither turned out to be relevant.) For each of those streaks, I calculated how much a prescient (read: lucky) fellow might have made betting $100 on the first game and entirely reinvesting.

(Quick sidenote: this is a fun exercise, but I’ll acknowledge it’s far from perfect. On a practical level, the data aren’t that trustworthy and the assumption that gambling lines proxy for probability estimates is shaky. More troublingly, on a theoretical level we would expect that the lines for the later games in a given streak shift some with a team’s wins in the first game(s). I don’t know enough to guess how much a given set of lines will jump around, but I suspect that this method overestimates the ex ante probability that a streak like this would occur. Also, if we guess that lines bounce around more in the early season, the odds on the Wizards and Bulls games probably dropped quite a bit, further underestimating how rare this streak is.)

Anyhow, as it turns out, only one team has had a three game streak that was as unlikely as these Sixers’ from a Vegas perspective. The post-lockout Wizards had a streak in April 2012 that is in some senses similar to the current Sixers’ streak. They were 14-46 and, with only six games to go, would presumably tank the shit out of the rest of the season for a shot at Anthony Davis. Instead, though, they beat the league-best Bulls on the road (at +675), Milwaukee at home (at +330), and followed it up by beating the eventual champion Heat in Miami (at +450). Riding them for those three games would have made you a profit of $18,228.75, which is just obscene—it’s an extra $5K (or 5000%) beyond the Sixers. As it turns out, they’d close out the season with three more wins, all as favorites, against two tanking teams and the Heat, who were presumably resting starters by then. If you’d kept piling on, those six games would have gotten you almost 47 grand…though it might have also merited an intervention.

There are only two other teams in the same ballpark as these Sixers and Wizards. The December 2007 Trailblazers picked up a profit of $10,956 and were buoyed by a win in Utah at +1200—apparently they had only won once previously on the road and were missing LaMarcus Aldridge. They were moderate underdogs the other two games (+170 and +215), but it was driven by that one game. The only other team above $10K (or even $8K) is the March 2012 Cavaliers at $10,508, who got there by winning at Denver and Oklahoma City.

So in nearly 5 years of data, we have one bigger streak and two that are in the neighborhood. As I said earlier, I think the Sixers’ streak is a little more unexpected than this method gives it credit for (even relative to other streaks), but by this method we figure that this is a once in five years occurrence, even if seems much odder than that right now.

All this said: even if I take the supposedly rational perspective that weird shit happens in small samples, it still doesn’t make me feel better about the Bulls’ blowing that lead last night, though. At least Rose is back: he’ll probably be fodder for posts later down the line.

*Andrew Wiggins is the presumptive #1 pick in the next draft. Andrew Wiggin is the real name of the protagonist of Ender’s Game. We can make this happen, people.