There's a link between Olivier Giroud and the financial crisis: both show the importance of reputation.
Yesterday, Giroud both missed horribly against Stoke and scored a good goal. To his detractors, the miss is evidence that Arsenal can't win the title with a second-rate striker. His fans, by contrast, look at the goal as evidence that he is good enough. What both are doing is seeing his performance through the prism of their priors; if you think Giroud is mediocre, you focus upon his shortcomings but if you like him you see his merits*.
This is how reputations are built. If you have a good reputation, a mix of the framing effect and confirmation bias means that people will read ambiguous evidence as compatible with that reputation. And if you have a bad reputation, they'll interpret the same ambiguous evidence as consistent with you being a duffer.
Reputations, though, are assets and assets can be exploited; as the old saying goes, "give a man a reputation as an early riser and he can sleep til noon."
One of the best examples of this in sport was Shane Warne's later career. His reputation as a magician so bamboozled batsman that he could take wickets even with ordinary straight balls.
A similar thing happens in music; a reputation as a great artist allows some performers to make self-indulgent albums: think of the more, cough, challenging music of Lou Reed, Prince or Scott Walker.
Conmen know this. One of the oldest frauds is the confidence trick. You win the confidence of the mark, for example, by repaying small loans with interest and then use this reputation as a good borrower to borrow a fortune and run off with it.
In Phishing for Phools, George Akerlof and Robert Shiller call this reputation mining:
If I have a reputation for selling beautiful ripe avocados I have an opportunity. I can sell you a mediocre avocado at the price you would pay for the perfect ripe one.
This, they say, is just what ratings agencies and investment banks did in the 00s. They used their good reputations to give high credit ratings to bad securities, thus passing off bad avocados as good ones.
Such behaviour isn't, though, confined to the financial sector. Luigi Zingales and colleagues have estimated that around one-in-seven US companies might be defrauding their shareholders by bosses' exploiting their reputations for trustworthiness.
The problem here, though, isn't simply outright dishonesty. Even wholly honest corporate bosses and fund managers use inflated reputations for excellence to earn big salaries and fees. And companies can profit - for a while - by selling inferior goods under a good brand name: its critics claim that this is what has happened to Cadbury's chocolate.
My point here is that exactly the same processes that we see among football supporters - a tendency to use cognitive biases such as Bayesian conservatism and the confirmation bias to build reputations which aren't always proportioned to evidence - also exist in economics, and sometimes with unpleasant effects. Economics and sport have much in common; in fact, I regard Matthew Syed and Ed Smith as our best economics journalists.
You might think all this just shows the dangers of having strong prior beliefs about people's competence and trustworthiness.
Maybe. But there are dangers in not having priors. There's a rare phenomenon known as the Capgras delusion, whereby people believe their close relatives have been replaced by impostors. When psychiatrists quiz such patients, they often reply along the lines that such events are indeed improbable but they must believe the "evidence" of their own eyes. To such patients, more faith in their priors - that people don't get replaced by impostors - would enhance their psychological health.
Good judgment requires us to hit the happy medium, putting neither too much nor too little weight upon our priors.It's because this is so darned difficult that costly cognitive biases are so ubiquitous.
* In fact Giroud's goals-per-game ratio, at 0.42, whilst less than that of the best strikers, is comparable to that of some lauded ones such as Carlos Tevez and Didier Drogba (and, ahem, Emmanuel Adebayor).
Meanwhile the economists have modelled Giroud (and all other footballers) as 3d squares that can all run at the same speed, are all the same height, all never cheat, all can run indefinitely. Both managers are tactically perfect and the game is fully deterministic. As both sides have the same boxes each game is a draw.
In 2008 I heard a couple of over-confident economists talking on Radio 4 about how they could improve any football team by applying the methods of high-finance. This was after the start of the credit crunch but before these numskulls had realised their world was a lie. I remember thinking, as they told the presenter that Ferguson was good but that they could do better, "thank god you guys are about to go off a cliff".
Happy to say I've not heard from them since.
Posted by: ben | September 13, 2015 at 02:49 PM
"Bayesian conservatism"? The conservatism cognitive bias - https://en.wikipedia.org/wiki/Conservatism_(belief_revision) - is a /departure/ from "Bayesianity"!
Posted by: phayes | September 13, 2015 at 03:15 PM
«call this reputation mining»
That is just a special case of the notion of asset stripping, or liquidating capital into profits.
Which are themselves manifestations of the Blissex Second Law :-), that all non-trivial frauds are variants of hidden under-depreciation of assets.
In this case the asset being under-depreciated is the reputation of the organization.
Posted by: Blissex | September 13, 2015 at 09:10 PM
As a bowler of straight breaks, I much appreciated the Shane Warne "slider" link.
Posted by: Luke | September 13, 2015 at 09:40 PM
Your example of Giroud rests on the assumption that good footballers are good at one thing, scoring goals, whereas they tend to be multi-skilled. While cricket has its all-rounders, the very fact that some players are known as such indicates that the possession of multiple skills is unusual. Shane Warne was only really good at one thing. A top footballer these days must be good at a number of things.
The decline of the "ultility player" in football since the 80s reflects the way that the game has shifted from a focus on a singular talent (good in the air, nippy) to the combination of skills, which has also had the effect of making team styles more variable in-game and players easier to rotate. His critics don't think Giroud is mediocre, but that he is incomplete - specifically lacking mid-range speed and quick feet in the box.
As Wenger has noted, Arsenal's problem is not a lack of cash to spend, but the absence of any "complete strikers", a la Ronadldo, on the market. His strategy is therefore one of combinations: Walcott + Sanchez, Giroud + Sanchez, Giroud + The Ox etc. The frustration of the fans is that sometimes the chance falls to the wrong player (Sanchez would probably have scored Giroud's first chance, but perhaps not his second).
The parallel here with economics is that investing in the reputation of an individual is risky because outstanding and complete talents are very rare. The more sensible approach is to look for synergistic combinations: a balanced portfolio, strength in depth, hedging etc. What matters is the team, as the hilarious antics of Chelsea are currently proving.
Posted by: Dave Timoney | September 14, 2015 at 05:26 PM
I don't usually find too much fault with FATE's posts, but I have to pick holes in his cricketing judgement. There are very few players at test level that cannot bat, and in the recent Ashes players with test centuries like Moeen Ali and Stuart Broad were coming in at 8, 9 or 10. Even batsmen can turn their arms over a bit, and it was an obligatory requirement for professionals playing in quality amateur/semi-pro leagues that they contribute at least 25-30 wickets a season even if they were never bowled in first-class cricket. It is also very rare these days that players cannot field adequately.
Compare this to football, where David Beckham made a very good living without being able to head, tackle or dribble very well, and his contribution was limited to set-pieces and long passes/crosses. There is also still a lot to be said for the cliche of a 'forward's tackle'.
Posted by: Igor Belanov | September 14, 2015 at 06:39 PM
Welcome back Chris, hope you had a good break. One of the key points here (i'll avoid the sports sidetrack!) is how the rating companies got away scott free despite showing their incompetence/complicity in over-rating junk CDO's (etc) that led to the financial crash. How they have any credibility is a source of confusion to me as they comprehensively demonstrated they are not meeting their basic requirements (although if R. Brooks can get a senior leadership role back after destroying a company either through illegality or incompetance, then anything goes these days...)
Posted by: scism | September 15, 2015 at 10:34 AM
@Igor, I bow to your superior cricketing knowledge, but I still maintain that you can get by in the summer game if you are outstanding at one thing but rubbish at everything else, whereas that hasn't been the case in football for some time.
Beckham had to move abroad to maximise his limited skills when the game changed, whereas a more all-round player like Giggs didn't. By the way, this thought struck me when I noticed Peter Crouch warming up for Stoke on Saturday. He never made it on to the pitch.
Posted by: Dave Timoney | September 15, 2015 at 04:43 PM
Crouch is a strange case. Slow and clumsy, yet technically pretty good- he scored some terrific goals with his feet. Oddest was the fact that he couldn't jump well and only got by in the air because he was so tall! (I also seem to have subconsciously ended his career!)
You're probably right that you can theoretically do well at cricket with a more narrow set of skills, just in practice this is rare. It's also interesting that many players have had ability in both football and cricket, and I'm sure we would still have the occasional double international if professional sports teams weren't so precious about their employees.
Posted by: Igor Belanov | September 15, 2015 at 05:08 PM