Friday, January 29, 2016

What a Crockett III

In the previous two pieces in this series we've been looking with scepticism at the playing record of Stephen Crockett, four-time Grand Prix champion and winner of numerous tournaments, recently retired from the circuit after questions were asked about the integrity of his results, yet nevertheless recently made controller of the very same Grand Prix.

It's only fair to observe that, following his own announcement of his retirement, he's made some effort, on his Facebook page, to provide an explanation of some of the more remarkable features of his record and to rebut charges of sandbagging. This is recorded in comments he himself has made to this posting. For the convenience of readers I have provided copies of these comments at the foot of this piece.

Mr Crockett plays down the possibility of sandbagging by observing that there is no meaningful financial incentive, in amateur circuit chess, in fixing one's own results to one's disadvantage.
The first thing to understand about the idea of 'sandbagging' is that chess would be one of the most pointless sports/games to do it in on the whole- as regrettably there just isn't any significant money to be made in it apart from at the top level (in the UK at least). The only way anyone could get a significant financial advantage from losing games on purpose would be if given one off bungs to lose a game (not common as there's no gambling on chess apart from at the very top), or for someone who was in the 'second tier'- i.e. a very strong/titled player who would struggle to win much in tournaments with super Grand Masters but could hope to take home a sizeable pot from being eligible for a grade restricted but still extremely strong- stars barred type tournament.
This is no doubt the case, although at the same time, it's not an argument that applies to a player of Mr Crockett's real or apparent strength. The point that follows, however, addresses his situation more directly.
Some people talk about graded sections lower down in chess and wonder if people may want to lose on purpose to keep in a lower section than they should do- but that's largely a red herring- there simply isn't any money in it (the costs of entering to and travelling to tournaments usually outweigh the prize money even if one is lucky enough to win the section not to mention the costs associated with all the fruitless trips you'd end up making while losing games!)- the only occasional time I could see this happening is if someone enters ay 2 or 3 local/big events which do have decent money on offer in a year and doesn't otherwise play a lot and could easily lose a few league games to 'manage' their grade.
This is the same argument as in the preceding passage - "there simply isn't any money in it" - and it's as true as it's irrelevant. Why would the motive of an amateur who cheats necessarily be money? Moreover, if our only motive is money, why would we play in the first place, since "there simply isn't any money in it"?

In reality people have all sorts of motivations for trying to win, be they some kind of glory, self-satisfaction, the admiration of others and many other things. Where those motivations exist for winning, they exist also for cheating. People do, in fact, cheat in amateur as well as professional competitions.

So if you see no reason for cheating, other than money, then you are really not looking very hard. As a rebuttal it doesn't even make it out of the starting gate.

The major part of Mr Crockett's explanation is more serious.

He argues that the reasons for the inconsistency in his results are health-related: that he suffers from depression and that this has caused him to have inconsistent results.

He also claims, in addressing the question of how he achieved a large number of scores of zero or close to zero, that he continued playing at events where he was feeling practically unable to play - partly because chess served as an addiction and partly because he'd already paid entry fees and hotel bills.

Key passages:
So how about inconsistency in some players play..why does it it sandbagging? In the vast majority of cases almost certainly not. Partly for the reasons outlined at the top of this thread e.g. lack of financial incentive, partly because of players pride and desire to increase their grade and playing strength. The reality is that there are a lot of reasons for wildly varying results- they vary by player and include age, health, other life priorities, tiredness, not being in right frame of mind/mental attitude, panic, time management, the strength of opponents they're playing/adjusting to a new level, type of opening/changing style..the list goes on..
In my own case there have been a combination of circumstances that have affected me and stopped my chess strength and grade progressing in the way i'd have liked so far. The overriding one is health related- not something I usually will talk about in a public forum but since August 2011 i've faced a near constant battle with depression/anxiety and linked mental health issues. I was diagnosed in October 2011 after a few rough months and some time of not being fit for work. I've been almost constantly on medication of varying levels since October 11. However, one of the things that also helped me cope and get over the worst of it was playing chess
That was the start of the 4 years of near constant chess playing- like an addiction- generally good for me but also far too much chess to be able to play consistently well all the time. Health wise I had a good period but then over the past four years have had several relapses involving periods of panic/depression/poor concentration and time of being unfit for work and having to raise/change medication doses until things settled down again. Sadly my chess form went along with the bounces and dips I was experiencing mentally and I had periods when I could barely concentrate one move ahead- and others when I was feeling good and confident and focused where I was able to play well and really relax at the board in the right way.
The long distance tournaments were often particularly tough as I was having sleep problems when not well which made the concentration problems which go with the illness even worse,but I was entering events weeks or months in advance when i didnt know how id feel- paying for hotels and I was addicted to the lifestyle so would keep on going and playing regardless of how I felt that weekend.
Now to a degree I am sympathetic to Mr Crockett. This is because I have considerable experience myself, both of depression and of playing chess with depression. Depression is a million miles from easy and your chess can, indeed, sometimes be crippled by it. It surely does make your results inconsistent.

But at the same time, I find it extremely unconvincing as an explanation for the bizarre pattern of his results. It's not an all-purpose explanation of everything.

It doesn't account, in reality, for the practical disappearance of mid-table results from Mr Crockett's record. It is as if depression always struck either in the most crippling manner or not at all. Is that really the way it works? Over a long period? Really?

You don't get weekends when you feel great in the morning but lousy in the evening, or when you can play at your best on Saturday but not at all on Sunday? You don't score fifty per cent or sixty percent or forty per cent on a regular basis simply out of happenstance?

A strange affliction indeed, that doesn't prevent the sufferer winning "a record breaking 48 tournament wins including five British titles" - an extraordinary record - but nevertheless seems almost to prevent its sufferer scoring between two and three out of five in a tournament. One that doesn't prevent the sufferer travelling or turning up to play, but does stop them being able to play any kind of decent chess.

The thing about depression is that while its effects can be manic in a variety of ways, they're not magic. The last thing that depression is, is convenient. Yet the way it's claimed to have worked out here has been really convenient. If you're Stephen Crockett.

So why has it apparently worked this way just for Stephen Crockett? Depression, unfortunately, is not rare, among the public as a whole or among chessplayers. Where are the other players who, over a long period, alternate score after score of close to 100% with score after score of close to zero? Why is this pattern of scores so unusual?

You can invoke health problems as an explanation of spotty results, sure. No problem. As an explanation for being at both ends of the bell curve at once, perhaps less so.

So is this really an adequate explanation of why somebody is able to win dozens of titles and yet is able to lose games in grade-deflating quantities - in other words, that everything just happened to work out to the sufferer's advantage?

- - -

Moreover, I'm a long way short of convinced that it can serve as an explanation for Mr Crockett's remarkable run of twenty-six rapidplay defeats and a draw in 2010-11, at a time when his standardplay results were apparently unaffected by anything that was happening to his health (and, by the way, prior to the period in which he points to depression as an explanation for unusual results).

That's a run of results that would be immensely unlikely even for the consistently weakest player at a tournament, which Mr Crockett was not. As it is, whichever way you calculate the odds, they're going to come out at millions-to-one against.

So how did that extraordinary sequence happen to occur?

- - -

Mind you, it's not really a question of whether I'm convinced or not. It's a question of whether other players are convinced. And it's a question of whether those players' representatives have undertaken a process of evaluating Mr Crockett's explanations that is sufficiently rigorous for them to be properly convinced.
- - -

Publically, at least, I've not seen word one from the ECF, who both operate the Grand Prix and grade the results that are at issue. I don't find this a satisfactory situation.

I am under no illusion that it would be straightforward to investigate any player's results and come up with a definitive conclusion as to whether or not they were obtained legitimately. If only life were that easy. On the other hand, it is not good for the integrity of English chess to ignore results which attract public comment in the way that these results have.

Other sports have often chosen to ignore suspicious results, or at very least have chosen to give that impression by keep any investigations that they've undertaken under wraps. Often as a result, they have run into serious credibility problems. A paragraph from a recent piece on alleged match-fixing in tennis is worth reproducing here.
If there is one lesson of the past year, as trust in football and athletics has been decimated by corruption scandals, it is that sports governing bodies must conduct their business in the light. The never apologise, never explain approach of the TIU does not inspire confidence.
Quite. Sports governing bodies must conduct their business in the light. Though, as the writer might have continued, by and large they don't.

Especially the ECF.

Claims of sandbagging in the lower sections of the chess Grand Prix are not, of course, remotely on the same level as accusations of doping cycling and athletics or match-fixing in cricket or in tennis. But the problem - the problem of the integrity of competitions - is the same. Where there are results about which there is reason to be suspicious, then you can't just ignore them. Because what are you going to do when they crop up again?

I would like to assume the ECF has not just ignored the reservations that people have expressed about this particular competitor's record. I would like to assume, too, that they have not left any complaints unanswered. I would like to assume further that they do, as a matter of routine and as best as they are able, look into the circumstances involved in any suspicious pattern of results that comes to their attention.

Otherwise it would be quite remarkable to have gone and made Stephen Crockett the controller of the Grand Prix.

Because it's one thing brushing a problem under the carpet. It would be quite another putting it on the mantelpiece.

[Entirely anonymous comments will not be accepted on this series of articles. Stephen Crockett was contacted more than once in connection with these pieces but failed to provide any information in reply.]
- - -


Anonymous said...

If you look at standard play results for both 2009-2010 and 2010-2011, you get a mixture of the poor, the moderate and prize winning. That's as you would expect for a frequent player. Also lots of draws in league games. But then his standard grade was 139 for both seasons and his rapidplay grade (six monthly) 138, 137 and then 133. July 2011 after the losing spell starting halfway through the British Rapidplay the rapid grade plummeted to 119 so 20 points worse than the standard play one.


Anonymous said...

Go on, plus all the ecf / bcc grades into a spreadsheet and output the stats.
His Facebook page kind of reminds me of Colm Daly's writing.


ejh said...

That'd be an interesting exercise (though not all the grades are available) but of course there's a problem with it if the grades are not to be relied on.

Re: Roger's point, of course there's nothing exceptional about having rapidplay grades and standardplay grades that differ a lot - have a look at my record, for example - except when the player concerned is winning lots of rapidplay games and competitions. That's where it starts getting whiffy.

Anonymous said...

An investigative statistic would be to look at typical out performance and under performance. So you take the tournament records of everyone playing more than x in a season and see what their best performance relative to their grade has been and their worst. I'd suspect the biggest variation would be of players in Opens, but Opens themselves are of variable strength depending on whether Keith, Mark and others show up. Someone winning an Under 120 with 5/5 might perhaps have faced a 110 field, so that's a 160 performance. Someone scoring 0/5 might only have faced a 100 or lower field, so that's 50 or worse.

Successive chess games by the same player aren't sufficiently independent to use the logic that if there's a 1 in 6 chance of losing one game, it's 1 in 36 of losing two. The mathematics of throwing unbiased dice and getting sixes doesn't apply.


ejh said...

I think they are, because you're not trying to locate an exact figure (in the sense that we know that the chances of throwing three consecutive sixes are 1/216) but to give an idea of the unlikelihood of a given sequence occurring. Of course there are various ways to perform the exercise and all of them are potentially useful.

Joe Skielnik said...

After winning a couple of "minors" I would expect a controller to seriously consider "promoting" a player to a higher section in subsequent events. Why would anyone object to that? Some tournaments now give prizes for "rating performance" (ie improvement against your current grade) rather than best score within a rating band. This suggests that the whole premise of the Grand Prix needs to be reviewed and possibly reorganised. FIDE grades are updated monthly so this situation could never arise anyway in most other countries. The ECF grading system is due for retirement.

Matthew said...

I'd like to see the ecf come out and support C or otherwise given the sheer amount of talk on this subject recently. They have a pretty shabby history of supporting their employees and volunteers, notably when arbiters had their reputations attacked in the British chess championships of 2012 by the then ECF president which potentially left them in a difficult position with their lives outside of chess too.

I'm sure C must have a life too outside chess if he is only an unpaid volunteer and it can't be very helpful to be left high and dry to face this continued innuendo and speculation by the organisation he is meant to represent.

ejh said...

I'd like to see the ecf come out and support C or otherwise

So would I, but if they do so I'd like to see that the evidence has actually been gone through properly.Of course had it been done properly back when the problem occurred, this situation wouldn't exist now.

After winning a couple of "minors" I would expect a controller to seriously consider "promoting" a player to a higher section in subsequent events.

So would I.

Anonymous said...

I disagree that the situation couldn't arise with an Elo based system. What would happen is that the Elo rating would rise and then plummet. Tournaments using the common practice of basing their entry limits on a particular list would still have to allow the entry of a plummeted star. It's only if you base entry requirements on maximum rating achieved in the past, that you avoid this problem and this could be done with the ECF system.

If you have an ailment which prevents you from playing well, should you declare yourself unfit and not play? It's gaming the system to use the poor results from these periods to retain an eligibility to enter restricted tournaments.


Andrew B. said...

Let's set aside the "sand-bagging" arguments. Let's suppose a hypothetical player has a habit, when a tournament has started badly, of losing interest/getting depressed and playing well below what he's capable of and losing most of the remaining games.

Wouldn't the honourable attitude for this player be: "Well, because of how grading works, I'm eligible for (say) the u120 event, but then I'd be playing against a field much weaker than me, so I'll enter the u160 event. After all, if things go wrong again I might just as well get 0/6 in the u160 as the u120"?

And not: "Well, those 0/6 events last month were pretty grim at the time, but at least they mean that now I can beat up another bunch of patzers, who entered this event in good faith expecting to meet other u120-standard players"?

Andrew B.

TC said...

There's evidence that C has entered quite a few sections that are higher than he needs to from the grading record though Andrew so am not sure its so straightforward as you suggest in this case. Also it seems a lot of the poor results (though not all) have indeed come in the higher sections he has played in. Besides the honourable attitude argument- (which a lot of chess players from my experience don't follow!!) if C entered tournaments and his record was publicly available to a tournament controller then by accepting his entry is not that an acceptance that he is eligible to compete under the same terms as others? A lot of people enter sections when they have had higher ratings previously and they are allowed to. Not all the responsibility can therefore rest with the individual player. If C paid his entry fee like everyone else and was accepted as an entrant then the other tournament players are also by default accepting his presence by entering the same tournament knowing he is playing. If organisers did not accept the entry then it couldnt happen. If other players didnt happily play in the event or complained then the organiser would need to consider that.

Abi said...

Andrew B - is your expectation that people should do this sufficient grounds to single out one individual like this and condemn him as some sort of cheat?????

Simon G - a rapidly becoming bored regular reader.. said...

When can we expect the next article about something significant in the chess world/one of the top players rather than this repetitive rubbish about some low level chess player as I think you have more than made your point?

ejh said...

When can we expect the next article about something significant in the chess world

Never, I should think. Being a regular reader you'll know I almost never write about anything significant.

If C paid his entry fee like everyone else and was accepted as an entrant then the other tournament players are also by default accepting his presence by entering the same tournament knowing he is playing.

Not sure I follow this. Do people usually know who else is competing when they enter a weekend tournament?

Jack Rudd said...

Simon G - the answer is "when one of the blog authors feels like writing about one of those subjects". That's the way this blog works.

TC said...

''If C paid his entry fee like everyone else and was accepted as an entrant then the other tournament players are also by default accepting his presence by entering the same tournament knowing he is playing.

Not sure I follow this. Do people usually know who else is competing when they enter a weekend tournament?''

Not in all cases but most will these days with on line entrant lists often published well in advance- plus they would see who is playing on the day and could complain to the controller if they didn't feel any entrant should be entitled to play.

ejh said...

Ok. I hope to get on to this subject before the series concludes, but while I think there's an argument for putting some onus on tournament organisers, I think the organisers themselves are likely to argue that it's not up to them to decide who is competiting legitimately and who is not.

ejh said...

(Comments off until morning again. See you then.)

ejh said...

Carry on. Part IV at 0955 GMT today.

Lee Bullock said...

Just a comment on C playing in some higher sections. Yes he did. But this is because my campaign against him was so intense he did it to prove a point. He did it to say oh look I sometimes play up. On the few occasions he did play up he lost virtually every game and quickly! Lost rating points and could say see I play up sometimes to me. He knows how dodgy it looks going up and down in rating and keeping his grade low also.

Anonymous said...

I suppose playing in an Open can be a subtle way of losing grading points. You don't even have to try to under perform if you aren't good enough to survive in elite company.