Football Officiating

Like everything ever written about officiating, this is sour grapes. I was motivated to think about it by the Michigan-Ohio State game, which featured some suspicious officiating. Having had some time to mull it over, I think J.T. probably got the first down. Or anyway, it was too close to call. I think Michigan kinda got screwed with a lot of the other calls in the game, but that one was okay.

However, the whole discussion highlighted again how I disagree with just about everything people say about sports.

  1. There’s a saying that I usually like quite a lot, “Never ascribe to malice that which is adequately explained by incompetence.” This is often applied to officiating along the lines of, “don’t ascribe to bias what could be explained by incompetence.” While I usually agree with this sentiment, in this case I think it’s actually quite misleading. It implies that officiating mistakes are based on either bias or incompetence, but not both. That completely mischaracterizes the relationship. Bias and incompetence aren’t opposites. They’re not directly related. However, incompetence opens the door for bias.

    This idea came from my experience with poker. In poker, if your strategy is simple and you know it very well, you can implement correctly even if you’re very tilted. For example, if I’m short-stacked in a tournament and playing a fold/all-in strategy based on a chart, I can execute that strategy even if I’m tilted. On the other hand, if you’re trying to play a very complex strategy that you haven’t completely mastered, even a slight amount of tilt will often cause you to make terrible decisions.

    Applying the idea to officiating, a highly trained, scrupulous official can likely override most bias he has towards the teams to usually make the correct calls. But a less skilled official is not constrained by the correct calls, because he really has no idea what they are. So in the absence of firm knowledge of the right call, his bias is more likely to shine through, because he basically has nothing else to go on.

  2. There’s no such thing as a hold.

    What I mean is, people seem to imagine if we could just define “holding” more cleverly, it could always be called correctly. Maybe some rules could be improved, but for most of this stuff there is no perfect definition. There’s no articulation of the rules that will A. Capture everything we think is holding, B. Exclude everything we don’t think is holding, and C. Be simple and straightforward enough to be called correctly in real-time 100% of the time.

    Football is played in close quarters by incredibly fast and aggressive men trying to get the upper hand in any way they can. These interactions are messy and most of the fouls aren’t really well-defined, nor can they be.

    Along those lines, I don’t really think of it as “hold or not,” I think of it more like a 1-10 scale of how much the action resembles our idea of a hold. This idea calls into questions a few things that get said all the time.

    First, “They’re gonna call that every time.” One of the announcers says this almost every game, but it’s clearly nonsense. Every football game I watch, 9s and 10s go uncalled, and 1s and 2s get called. There’s nothing that gets called “every time.”

    Second, “That’s a great call.” This is often said in reference to something that was called a foul, which didn’t especially appear to be a foul in real-time, but with benefit of replay you can kind of see a plausible foul. This is often cited as a “great call” because the official picked up on something non-obvious. In my book, this is a run-of-the-mill bad call, because it’s basically calling a foul on a 3. It’s inconsistent application of the rules, because if such things were called consistently there’d be multiple fouls every play.

    Third, when outside observers review a game to look for bias, they typically look for “howlers”: completely unjustifiable calls. Well, you’re not really going to find that, because almost every play in the whole game is ambiguous in some way. What bias would look like, if it’s there, is stuff like 3s getting called on one team and 7s going uncalled against the other. When they find this, they typically say something like, “Well there’s not really evidence of bias because these calls are all individually justifiable.” What were you expecting to find?

  3. “They could have won if they played better.”

    I’m surprised how often smart people say some version of this. This is generally a way of one-upping someone who’s complaining about the officiating. The thing is, this is true of literally every team that ever lost a football game. If they had played sufficiently better, they could have won.

    What I don’t like about the statement is it’s misleading. It seems to be saying, they didn’t earn the right to c.omplain about the officiating because they didn’t do x, y, or z. But what it’s really saying, logically, is that it’s never valid to complain about the officiating. Which is a fine thing to say, but why not just say that then?

    I think the motivation comes from the knowledge that, in your personal life, it’s generally more effective to focus on what you can control than complain about things outside your control.

  4. People often suggest we shouldn’t worry about the officiating too much because it usually doesn’t determine the outcome of the game. In my view, bad calls determine the outcome of the game all the time, but it’s still not clear we should really care very much.

    Football is entertainment. It also has a tremendous amount of luck in various forms: the bounce of the ball, the interaction of play calls, etc. Ref error is just another form of luck. It’s only a problem if the perception of unfairness damages the entertainment value.

    Along those lines, I think the trend towards more replay reviews is a mistake. It hurts the entertainment value by slowing down the game. Fairness isn’t the priority. The perception of fairness only matters insofar as it makes the game more entertaining. In moderation, bad calls can even make the game more fun, like a wrestler hitting the opponent with a chair when the ref’s back is turned.

  5. For those reasons, I don’t see officiating reform as a very important priority. There are, however, some interesting reasons that make it very difficult to achieve.

    First, there’s a collective action problem. As long as the officiating is merely incompetent and not crooked, it shouldn’t benefit one team over another in the long-term. Therefore, no one is incentivized to expend the resources to make it better.

    The only time someone gets worked up to make changes is when they get screwed over, but then they lack credibility because their complains are seen (rightly) as sour grapes. So in all, it is nearly impossible to have both motivation and credibility at the same time.

    The only exception is when the person who gets screwed over has such high status that they can take the credibility hit of sour grapes and still press forward. This is how replay got instituted in the Big 10 and eventually all of college football. Joe Paterno got pissed when Penn State ate a bunch of bad calls in one game and moved to add replay. He was able to succeed because he was arguably the most revered figure in college football (this was pre-scandal), so his cache trumped the sour grapes issue.



The Phases of Aces

Not once but twice in my session the other day someone had the chance to go all in pre-flop against me with pocket aces, and declined. What’s strange about this is that aces are the nuts pre-flop. If you’re not going to get your money in with the best hand, what exactly are you trying to do at the poker table?

It’s actually a little more complicated than that. In Omaha, aces aren’t in fact a hand, but a range of hands. Ace-ace-jack-ten double-suited is a much different – and better – hand than ace-ace-seven-deuce with no suits. Additionally, aces aren’t nearly as powerful in Omaha as in Hold’em. In Hold’em aces are about 85% to win against a random hand. In Omaha it’s only 65%. Nonetheless, aces are still favored against any non-aces starting hand. So why didn’t they go all in?

I’ve found there are certain patterns in how people play their aces in Omaha. At first, they go nuts with them. This is probably because most people come to Omaha from Hold’em, where aces are incredibly powerful. They raise pre-flop, raise post-flop, and keep betting until all the money’s in. This may work for a little while, but inevitably it costs them some big pots, because it’s not really a good strategy in Omaha. On the flop an overpair with nothing to go with it is often nearly worthless in Omaha, especially against multiple opponents. After losing a few pots where they overplay aces, novices soon learn to be more circumspect. Often, they’re helped along by advice from an old-timer that aces are nothing special in Omaha. Not exactly true, but common wisdom nonetheless.

In the next phase of their aces development, they do an about face, and play them incredibly passively. They almost never raise them unless maybe they can get all in – and sometimes, like I was saying, not even then. Many players are very concerned that if they raise pre-flop they will “advertise” their aces, so they play passively to keep them concealed.

And this is where many players’ development with aces stops. You might think that given a long enough time frame, feedback would eventually push players towards an optimal strategy. But in poker at least there appear to be ruts – places where quirks of psychology or the game itself push players into bad strategies that they can’t escape from. Playing aces passively is a very common example of this. It’s a well-known facet of our psychology that negative experiences have stronger and longer-lasting effects than corresponding positive experiences. Many players fixate on a few very negative experience with aces, causing them to undervalue them from then on.

So what’s a better strategy with aces? Well it is true that effectively revealing your cards early in the hand with a lot of money left to play for is a huge problem, one that will be exploited ruthlessly by good opponents. For that reason raising with aces and no other hands is a very bad strategy. As is often the case in poker though, the best strategies turn out to be quite aggressive. Rather than concealing your aces by playing them passively, it’s usually a better idea to play a lot of hands besides aces aggressively. This disguises your hand while keeping raising in your arsenal, which is very powerful. You certainly want to be able to put more money in the pot when you have a great hand.


It’s the time of year for my annual pilgrimage to Vegas. Last year I was very excited to play because I felt I had come so far as a poker player, then proceeded to lose my first ten sessions. Even though I eventually bounced back and ended up making money on the trip, it was a rough couple weeks.

This year I’m coming in with different expectations. I don’t really see Vegas as a money-maker anymore. In general, I’ve found the games are better in Detroit. Now I see the value in the trip more in things like measuring myself against other pros, stoking my ambition, and making connections. I’m taking a more relaxed approach to winning and losing.

I’ve only played a few sessions so far, but I’ve already booked a couple wins, so I’m ahead of last year at least. It would be tempting to say my new approach is working, but as always in poker, luck is the biggest factor in the short term.

Conspiracy Theories

It’s been suggested that conspiracy theories are an attempt to match causes to effects. It’s hard to believe that random, mundane events could have earth-shattering consequences. It’s tempting to supply causes that match the events in their scope and complexity.

A similar thing happens in games. When a decision has disastrous consequences, it’s tempting to cast it as a blunder retrospectively, even if it was only a minor mistake, or not a mistake at all.

An example is the Seattle Seahawks’ decision to pass from the goalline in the Super Bowl. Memorably, this decision resulted in an interception that cost Seattle the game. There was immediately an outcry on social media with many fans and commentators blaming Seattle’s loss on their decision to pass. Yet, sober analysis indicated that the decision may have been a good one, or, at any rate, can’t really have been that bad. In the 2014 season, passes from the 1-yard line had resulted in 66 touchdowns and no interceptions before the one in the Super Bowl. Nonetheless, there was a need to make the decision to pass a blunder commensurate with the disastrous outcome.

Chances are, your biggest mistakes aren’t the ones you remember.

A Coup Endgame

Previously I wrote about a strategy for heads-up Coup, but it’s a better game with more than two players and that’s how most people play. The rules can be found here.

I got into an interesting endgame situation the other night. The game had started out with many players, but was down to just two, me and my brother. I have two coins and one influence remaining. My face down card is an assassin, which he knows is likely the case based on previous turns. My brother also has one card remaining, which I believe is probably an ambassador. He has one coin. I didn’t record which other cards were exposed, but that should be an important piece of the puzzle as well. It’s my turn.

I take one coin, intending to assassinate him next turn. He uses his ambassador to trade cards with the court deck. I already thought he had an ambassador, so I don’t challenge. Naturally I watch closely as he draws his two cards. He quickly and, it seems to me, confidently, chooses one of the new cards and shuffles the other two back in.

At this point I realize I need to seriously consider changing my plan. Since he believes I have an assassin he clearly recognizes my intention to assassinate him on my turn. It seems pretty clear that he decided to dig for a contessa and my read is that he found it. If the top two cards hadn’t contained a contessa, he might have looked disappointed. What’s even more telling though is his timing. If he sees a contessa his choice is obvious since that was his plan, but if neither card was a contessa, he’d probably need to stop and think about a new plan for winning the game. So I think it’s very likely that he has a contessa and if I try to assasinate he will block.

So what else can I do? I can take foreign aid (2 coins) twice and easily win the race to coup him. This plan can be foiled if he has a duke, assassin, or captain. Nonetheless, I’m pretty confident in my read and take foreign aid. He pauses to think for awhile and takes foreign aid as well. I take foreign aid again, bringing myself up to 7 coins. He tries to assassinate me but I assume this is a desperation bluff and challenge, and he indeed has the contessa.

In this case it worked out but how good was my strategy? It seems weird to open myself up to losing to 3 different cards (duke, assassin, captain) with the foreign aid plan when the obvious assassination plan would only lose to contessa. With no other cards revealed, he finds a contessa about 45% of the time, so I can guarantee myself a 55% chance of winning just by assassinating and challenging.

The simple assassination strategy has a 55% chance of winning. What if I went for an all-foreign aid strategy? Even if he does nothing to prepare for this, it would still do worse than the assassination strategy, since when he fails to find a contessa he usually finds one of the three other cards that foil the foreign aid plan, and that happens more than half the time. If he knew I was going to foreign aid it would be even worse and he could beat me almost every time by finding a captain, duke, or assassin.

Clearly all-assassination is better than all-foreign aid, but could a mixed strategy be better than either? I don’t think so, because as we’ve seen, even if he goes all-out to stop the assassination plan, foreign aid still has a lower expectation. So it seems to me with best play by both sides, I have to assassinate, and the game simply comes down to whether he finds a contessa. In the actual game I went with a read that allowed me to deviate wildly from the generally correct strategy.


“In actual fact I don’t have any clear preferences in chess. I do what I think circumstances require of me – I attack, defend or go into the endgame. Having preferences means having weaknesses.” – Magnus Carlsen

“The true master has no preferences, only abilities.” – Matt Sperling

Is a style anything other than a weakness? When working on your game, should style be cultivated or eradicated?

In poker we characterize players along two axes, tight-loose and passive-aggressive. The first describes how many hands you play, a lot or a little. The second describes how you play them, passively (mostly checking or calling) or aggressively (mostly betting or raising). It’s generally accepted that winning poker strategies tend to be aggressive. The two acknowledged winning styles are TAG and LAG, tight-aggressive and loose-aggressive. Historically there’s been a lot of debate in poker circles over which style is superior. My own suspicion is that tighter and looser strategies would have similar expectation against a good response; what playing looser really does is make the game more difficult for you and your opponents. It makes sense if you are better equipped to handle those difficulties.

Along those lines, style is valuable as a way of acknowledging weaknesses, both your own and your opponents’. The idea of “game theory optimal” play doesn’t really leave any room for a personal style, but you should keep in mind that you’re a human, playing against other humans. It’s not feasible to play optimally in all situations. Style could be seen as an attempt to prune down the game tree and focus on a particular branch to master.

Even if it doesn’t play into poker theory, style could be very valuable pragmatically. A boxer may switch to a southpaw stance to throw off his opponent. Few would argue that southpaw is fundamentally better than orthodox, it’s just a way of mixing things up, of searching for an edge. The ideal poker player would be proficient in many styles and be capable of throwing all of them at an opponent to find which bothers him the most. It would be a good sign for your game if different opponents would describe your style very differently. In this sense style is tied to nettlesomeness. It’s about having different routes to get under your opponent’s skin.

If used correctly, style isn’t a weakness, but rather a way of understanding and manipulating weaknesses, both yours and your opponent’s.

Poker in Lake Wobegon

Garrison Keillor described Lake Wobegon as a place where “all the children are above average.” Although Lake Wobegon is fictional, psychologists have found that the “Lake Wobegon effect” is very real. In surveys most people consider themselves above average drivers. When married couples are asked how much they contribute to household chores, their answers always sum to more than 100%.

Poker has its own version of the Lake Wobegon effect. Nearly every poker player believes himself to be an excellent player beset with bad luck. This belief is so universal that, if you took their word for it, you would have to conclude that all poker players are unlucky. But poker is a zero-sum game; one player’s bad luck is always someone else’s good luck. Clearly all poker players can’t be unlucky any more than all children can be above average.

Why does everyone feel unlucky? It seems to be tied to the need to feel exceptional. If they took their results at face value, most poker players would be forced to accept that they’re not any better than their opponents. This turns out to be far more painful than finding ways to see yourself as unlucky. Since poker is a complex game with many unexpected twists and turns, this is never very hard to do.

One common tactic is focusing on the point in the hand where you were ahead, ignoring all the other points. For example, a player might get all-in pre-flop as a slight dog, hit a great flop that makes him a big favorite, only to end up losing on the river. Of course from the strategic point of view you should only be concerned about the odds when the money goes in, but this rarely stops the player from complaining about how unlucky he was to lose after being so far ahead on the flop.

The “bad beat story” is a classic poker archetype. There’s always someone eager to explain how unlucky he got, usually to the intense boredom of everyone else. When I hear a bad beat story about a hand that I witnessed, I almost always notice that the player has changed the hand – sometimes subtly, sometimes drastically – to make it seem more unlucky than it actually was. The changes often multiply in the retelling: the opponent goes from 6 outs, to 3 outs, to 1 out…

There are no “good beat” stories. Even when players recognize their instances of good luck, they tend to see them as well-deserved but inadequate payback for their constant bad luck. Or they find a way to explain that it really wasn’t good luck at all. “Sure, he was ahead when the money went in, but he can’t expect to win playing like that.”

Some players are so wedded to the idea of their own bad luck that it almost seems like they play poker purely for the purpose of feeling aggrieved. For them, poker becomes a sort of cosmic game of chicken, where they are determined to test the limits of the universe’s malevolence towards them. This can easily become a self-fulfilling prophecy as they play worse and worse, leading to worse results which must be explained by even more bad luck.

The Play Every Hand Game

To deflect charges of being “on lockdown” I started up a challenge with two other players at the table to see who could play more hands. I have the reputation of being a tight player and it’s true that I play less hands than a lot of my opponents. Unfortunately, I think this is just the best way to play 10-handed PLO. It’s often not possible to turn a profit with garbage hands, even if you have a big edge post-flop. But a breath of fresh air is always nice and if nothing else it makes the game more fun. I don’t particularly like folding, it just happens to be the right play pretty often.

Like the trips-and-quads game, the no folding game adds a new wrinkle and it can be fun to figure out the best strategies for new rules. I figured I would be giving away some EV, but as things went on I started to wonder if this rule actually helped me. I felt like I might have been adjusting to the new rules better than my opponents. For example, marginal multi-way hands tend to be better candidates for aggression pre-flop, while nutty one-way hands should often be played passively. Something like 9763 double-suited is often a 3bet-or-fold hand. If you’re determined to play every hand, you can 3bet it. On the other hand, something like AT83 single-suited works better as a call. If you want to stay out of trouble you can, and maybe should, fold both of these hands, but if you are committed to playing a lot of hands it helps to know which to raise and which to call.

It also didn’t hurt that I immediately started god moding (straight flush vs. nut flush, etc.). At least one old man quit the game in bewilderment and terror. At the end of the session I was left wondering if I should open up my game more. At the very least, if you’re comfortable in the games you’re playing and can take a losing session or two, I think it’s a good idea to take a session now and then to drastically change your game and see what happens. Even if it doesn’t work the perspective you gain is likely to be very valuable.

Chess at the Poker Table

I first got into poker by way of chess, and while I don’t play chess tournaments any more, I still teach a few chess lessons and play online – sometimes at the poker table. I’ve got an app on my phone and sometimes when the action’s a little slow…

This is problematic in all the ways you might imagine. I’ve been known to fold a marginal hand so I don’t run out of time in my chess game. It’s also possible that being distracted causes me to play worse. A friend of mine wrote an article on this topic, as he reminds me every time he walks up to the table and I’m on my phone, which is often. He’s right, of course, that it’s better to stay focused, but playing live poker is like walking along a bridge with steep drops on either side. One side is intolerable boredom and the other is playing so bad you’re not profitable. I’m not saying I’m square in the middle of the bridge, but I haven’t fallen off yet.

Once a Canadian kid named Jordan noticed me playing on my phone. We struck up a conversation about chess and pretty soon he had downloaded the app and we were about to play. As if from nowhere, Kip floated into the conversation.

“How much ya playing for?”


“Mind if I book some action?”

Kip is one of those inveterate hustlers who has made his living gambling on one thing or another for decades. He managed to book $25 of side action on me. The game wasn’t very close.

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.c3 Nf6 5.d3 Ng4 6.0-0 0-0 7.d4 Bb6 8.Bg5 Qe8 9.h3 h6 10.Bh4 g5 11.Nxg5 hxg5 12.Qxg4 exd4 13.Qxg5+ Kh7 14.Qf5+ Kg8 15.Qg6+ Kh8 16.Bf6#

48 Hours

Last week I went to a new home game for the first time. I felt like I was in the scene from Goodfellas where Pauli slices garlic for tomato sauce with a razor blade: old guys drinking, cooking, and eating. After hanging out for about an hour we all go into a little room with just a poker table and chairs and start playing. Compared to the smoky, dreary casino, this really feels like what poker is supposed to be. Shooting the shit, having a whiskey, playing cards.

I immediately lose my first two buy-ins, but it’s only $300 at a time, at least at the beginning of the night, so I don’t really mind. Anyway if I lose the first night that should increase my chances of getting invited back. But I start to make a comeback and after a few hours I’m the big winner in the game, sitting with about $5000. It’s always bad form to “hit and run,” but especially since this is my first time at a home game, I can’t leave. More and more players drop out until we’re down to four. I’m hoping someone else will quit so I can go home and sleep, but no one does. We bump up the stakes and keep playing. At one point I get all-in with the best hand for just about all the money at the table, but I lose, then lose some more, and am back to nearly break even. I make another comeback and am up about $3000 when we finally quit at 9:00 a.m.

I’m planning on playing my regular 5/10 PLO game at the casino at noon, and at this point I’m not feeling tired, so I decide to just power through. I get a bagel and coffee at a nearby deli and hang out and read for awhile. I drive to the casino about an hour before the game starts and try to nap a little in my car. At this point I’m starting to get a distinct underwater feeling that I remember from staying up all night in school. I consider just going home and sleeping, but by now the game’s about to start and I don’t want to lose my spot on the list.

When I go in, alongside the usual 5/10 game, there’s a 5/10/20/75 game that’s been going all night with some of the biggest action players. I hesitate for a few minutes, but there’s really no choice. It’s too good of a spot. I jump into the bigger game and immediately lose $4000 in the first orbit. I’m thinking maybe this wasn’t such a good idea, but at least I’m feeling more awake. On my second buy-in I start to win a few pots. I set over set someone for $6000, then get all-in pre-flop four way with aces and hold up. Pretty soon the action players start dropping out and we’re just playing shorthanded with regulars. In contrast to the home game, this isn’t really a spot where I’m obligated to keep playing. In any case, I get the sense everyone is about ready to pack it in. I quit at 6:00 and go home and sleep until the next morning.