Money in chess

[WBuchanan]

Got bored in the library, so decided to compile some data on variables for congresses going back to 2010.

Ran a regression (with dependent variable being number of entries in Open) and included prizefund, first prize, grading prizes, distance from central belt and number of titled players (FM/IM/GM only) as independent variables. Albeit with a small sample size (n=50), and probably missing plenty of variables, the statistically significant variables are the total prize fund (p<0.01) and the number of titled players (p<0.02). Either players entering opens care about the number of titled players; or the number of titled players is acting as a proxy for another factor.

Hi Jonathan. Interesting idea. Even if you had more data, how could you draw any conclusion regarding cause and effect though? Could be the other way around - presumably the titled players would care about the number of entries, prize money etc. Most likely there is an interdependence (with most people acting on their own experience of how big the tournament usually is) making them impossible to separate just using stats.

What did you get for the possible effect of the grading prizes? Although drawing a conclusion on this would be open to the same objection in that a larger entry with more titled players would create more of a need for the grading prizes.

Asking the players what makes them more likely to enter (titled and mortal) might shed some light (maybe such a survey has been done before?). Then again, it might not..
Cheers

[Craig Pritchett]

I think I can clear up the influence of one key variable! Big money prizes attract almost all titled players, most of the time, and ambitious, higher-rated untitled players. Even though there aren't really very many (if any) truly (or at least solely) "professional players" in Scottish chess, they will still turn out most for big prizes. These have, of course, slumped dramatically in real value terms since the heyday of the (often very well-sponsored) UK weekend circuit in the 1970s and perhaps also through the 1980s.

It would be a more interesting use of stats to get a truer fix on that decline. It's easy to do (but might take a bit of time to do properly) on the basis of data amassed historically by the Treasury as well as other open access data provided by various other government and non-government bodies.

Empirical evidence!? Speak to the likes of Mark Hebden and Keith Arkell (the latter recently told me that he can hardly remember when he last could be bothered to play in a UK weekender), both of whom will confirm that weekend events no longer matter so much to them as the potential for economic gain has declined.

[Jonathan Edwards]

Albeit with a small sample size (n=50), and probably missing plenty of variables, the statistically significant variables are the total prize fund (p<0.01) and the number of titled players (p<0.02). Either players entering opens care about the number of titled players; or the number of titled players is acting as a proxy for another factor.

Or your sample size is misleading small.
Disraeli would approve.

Perhaps. Although I mentioned the small sample size... t-statistics are meant for exactly this - small sample sizes! In fact, when d.f.>30 (here it is 49), it converges to standard normal distribution.

Anyway, surely it's quite intuitive that titled players have a positive effect on the number of entries in an open.

Hi Jonathan. Interesting idea. Even if you had more data, how could you draw any conclusion regarding cause and effect though? Could be the other way around - presumably the titled players would care about the number of entries, prize money etc. Most likely there is an interdependence (with most people acting on their own experience of how big the tournament usually is) making them impossible to separate just using stats.

Hi Walter. That's the reason for using regression analysis - determines causation and not correlation. Of course, more data and variables would increase the power of the test. It would be possible to include lagged variables from the previous year to account for how big the tournament is; or even use panel data, but it would take such a long time to find data for the past 10 years or so.

What did you get for the possible effect of the grading prizes? Although drawing a conclusion on this would be open to the same objection in that a larger entry with more titled players would create more of a need for the grading prizes.

The effect of grading prizes was slightly positive (but statistically insignificant). Interesting point that there may be simultaneous causality. I'd love to be more rigorous if it was worth the time. It was more just to demonstrate that encouraging titled players to enter congresses might be a good thing.

[WBuchanan]

Hi again, you say

"Hi Walter. That's the reason for using regression analysis - determines causation and not correlation."

I don't think this is true Jonathan. Regression gives the best parameters should a correlation exist (and a measurement of the statistical strength of a correlation), but cause and effect are not included; they are always a matter of interpretation. That's always the most important part of any statistics! Cheers

[Adam Bremner]

Anyway, surely it's quite intuitive that titled players have a positive effect on the number of entries in an open.

I think this really depends on who they are. I accept that a 2500 will provide a reasonable lure, but does a 2300 really provide the same? I can only speak for myself when I say no, and I hate stats too much to try and test this.

It feels like Craig's point is where the truth lies, if the most statistically significant event is prize money. The big prizes attract everyone and that includes the majority of the titled guys. It would take a brave congress organiser to increase prizes in the hope that they would get more entries though.

[Phil Thomas]

Hi again, you say

"Hi Walter. That's the reason for using regression analysis - determines causation and not correlation."

I don't think this is true Jonathan. Regression gives the best parameters should a correlation exist (and a measurement of the statistical strength of a correlation), but cause and effect are not included; they are always a matter of interpretation. That's always the most important part of any statistics! Cheers

Walter,
totally agree with your comment about interpretation - for any statistical work. Classic example in my mind is that one can look at the amount of damage done in a fire and correlate with the number of fire engines attending. If not careful you would conclude that the fire engines are causing damage.

However I strongly applaud the attempt at analysis by Jonathan. By definition he is not handling data that is not from a controlled experiment and there is every chance that key controlling parameters have not even been measured.

My work based experience with that kind of data is that you start off talking about the strongest correlations and therefore start off by telling the customer what he already knows.

In this specific case
Strongest result links high prize funds with large number of entries. Makes sense to me - in general prize funds come from entry fees. Tournament tend to budget from size of last year's entry which should be a good predictor of this year's entry.

Second strongest result - if I paraphrase it correctly - titled players are more likely to enter when field is large. This would be necessarily so if titled players are driven by the same factors as everybody else.

Jonathan,
what would be really interesting would be the next strongest findings from your analysis.
Especially interesting if they are non intuitive results.

Talking about controlling factors not listed.
Great playing room for the Edinburgh congress this year makes me more likely to play again next. For me it scores for view from window, acoustics, comfortable ambient temperature without draughts, comfortable chairs. Four things that can't go into any statistical modelling without incredibly detailed questionnaires - which most of us wouldn't bother to fill in anyway.

[Matthew Turner]

Kevin raises some interesting ideas at the beginning of this thread.

1. Free entries to titled players

I have no strong feeling over whether this is the right thing to do or not. However, I would say that I am personally very grateful when tournaments offer free entries. It is a relatively small amount of money and I agree with the majority of contributors who've said it is really up to organisers to decide what they want to do with their own congress.

2. Prize money

Over the past 15 - 20 years the prize money at the Scottish congresses has really declined and I sense that there is a strong feeling that this doesn't really matter (because it was often English GMs coming across the border and taking the money). I think this is a really misguided perspective - I think Scottish chess has lost a lot from not having the fairly regular £500, or even £750 first prizes.

3. Entry Fees

Kevin suggested that entry fees should rise 20% and they would still be dirt cheap. That is a tough one, when you've got travelling and accommodation expense to account for as well then entry fees are relatively cheap in the overall scheme or things. However, for a significant number of players I would suggest entry fees are far from dirt cheap.

4. Spending priorities

Kevin suggested that Chess Scotland should spend no money on adults and everything on junior chess. I have a lot of sympathy for this point of view, however I think it is quite a lot more complicated than Kevin suggests. Firstly, I think there is a third main area of expenditure that may become increasing significant to federations, namely seniors' chess. After all we are talking about spending £5,500 on the Olympiad team where the prospects of a medal are zero and a lot of money will be spent on international junior chess (mainly by the parents) and the realistic prospects of a medal are probably only a fraction above zero. On the other hand, Craig Pritchett represented Scotland at the European Seniors (presumably with relatively modest expenditure) and was seeded 4 or 5.

When we talk about junior chess (in federation terms) we are often talking about World and European Championships. As you will see on the thread started by Jacqui Thomas the cost of these events is colossal and the whole of Chess Scotland's resources could soon be eaten up. I would suggest that if you want to prioritise junior chess you need to think clearly about the most effective ways of spending the money. As Robin Moore points out the Commonwealth Championships in Glasgow is surely a much more economic way of giving our best juniors international experience than jetting them half way around the world. I appreciate that this is an unusual situation, but in normal years there are lots of big European Open which are easy to get to, better experience and vastly cheaper than World and European Championships.

[robin moore]

Matt,

Just to say I am not against sending juniors to overseas events especially if they are selected to receive the funded place. However, as has been said on the thread you mentioned, the Commonwealth and Glorney are great events based in Scotland in 2014. I am sure they will be strongly supported by our top juniors.

There are great events that would appeal to juniors outwith the World Youth, Euroyouth and EU Youth but no funded place would be available and parents would find it difficult to take their children out of school during term time other than to the pinnacle events.

[Matthew Turner]

Robin,
I think the issue is what Chess Scotland should prioritise with its limited funds. If parents want to go to Brazil (for example) to represent Scotland at the World Youth Championships then of course they should be encouraged to do so. There may be all sorts of benefits from this sort of experience. However, that does not necessarily mean that Chess Scotland should be subsidising the costs. There may be much more 'efficient' ways of using limited resources and I agree that the Commonwealth Championships in Glasgow is an opportunity that Scottish Junior Chess certainly shouldn't pass up.
Matt

[Patrick McGovern]

old one but may be appropriate here;
How to make a million at chess?
Start off with 2 million!