Black cat in a Ballpark

Random enough headline? I picked up a book called “How to Find a Black Cat in a Coal Cellar” after encouraging review by James Butler of Betfair Pro Trader. And the review didn’t lie, it is a good book and I can wholeheartedly recommend it. And as mentioned by James in his review, while it is written from a viewpoint of analysing tipping services nothing prevents one from using the same ideas when analysing homegrown systems and strategies. I plan to use these tests for my own systems later on but as a practice run I will begin with analysing a betting record I have from last year.

At the beginning of my betting life I ran a tipster proofing blog in Finnish. One of the tests and the longest running one was for a service called Sports Picks Buffet. Service collected tips from several sources and then chose best bets based on ratio between opinions of different tipsters. Service alternated between seasons with MLB and NBA and during my test all tips were for baseball. Buffet was ran under a pseudonym and fake Phd credentials which person behind the system came clean about a short while after my test concluded. Currently the service is run under a presumably correct name and shortened name of The Picks Buffet.

First. lets look at basic info about the betting record. There were total of 336 bets over roughly a three month period and all were baseball moneyline bets. During the test period meagre profit of 6.38 points or return on turnover of 1.90% was achieved. Strikerate during the test was 60.12% and average odds were 1.71. In the chart below you can see how bets unfolded.

As you can see, there was an healthy upwards trend in the results, which was followed by a decline and a new climb. Profits never rose to any extreme heights.

How about statistical significance? Testing this was biggest takeaway from the book I mentioned at the beginning of the post. Anyway, for these bets we can calculate a p value of 34.04% meaning that there is over one third of a chance that results above are purely due to chance alone. Black cat book recommends a conservative limit of 0.01% and when value should be below of that for results to be considered statistically significant I think it is safe to say that these results are not.

If we were to draw a normal distribution with average results above we would get a figure as below. But ofcourse there were many bets above and below of 1.71 that was the average. Betting with odds of 1.71 and fair odds of 1.66 there would still be a 26% chance of negative return.

(Sport) Picks Buffet is priced at pretty expensive 125$ per month and I don’t think these results quite justify that kind of a pricetag. Especially when there is service providing basically a same kind of service for cheaper. I haven’t personally tried this, but based on info on their site it could be used in similar way. So to me Scamdicappers seems to offers the same thing, only for multiple US sports and for 50$ per month.


Progress in my hunt for a good run

Now that I finally found  a way to calculate if a race was good, bad or merely OK. I took all races run between 1.10.2012 – 30.9.2014 and looked at last run each runner had had and determined if it was good bad or ok.

Now I was able to calculate if my method was up to anything, reasoning is that if last run was a good one then that should improve the odds of winning for that horse. And it actully did, from the table below you can see how performance in previous race affected strike rate.


None of the the above made any profits but at this point I am more interested in ways that let me forecast the winner and I am happy with anything that has a better strikerate than choosing randomly. And in races in question betting randomly one would have achieved strikerate of approximately 10%. My good runs last time out perform better than that and as importantly, races deemed bad perform worse.

I didn’t expect to find profits by using only one indicator so I was surprised that Hunter Chases and NH Flats combined made almost 600 points of profit over two years using just this one indicator. And that is over almost thousand races with roughly 450 winners. Not too shabby I’d say.

What is a good run?

Earlier I wrote about my plan to add a new kind of rating called a Suitability Score. And for that I needed to know if run horse had was a good one even if it didn’t finish first.

Initially I used a concept from Peter Mays book, but it was obviusly only for flat races as the system broke down when distances went further and  this lead to chases and hurdles having a lot more good runs compared to flat races.

After staring at lengths lost for flat and jumps for a while I came to conclusion that one-size-fits all solution wont work here and I also started thinking about going, it has to have effect on lengths a horse lost by, or atleast that is the hypothesis at the moment.

I was surprised that there seems to be greater variance between race distances in flat races compared to jump races, as can be seen from two charts below which describe average lengths lost per BSP for different distances in furlongs.

As I have recently dabbled with Artificial Neural Networks I decided to see if they could be of help in determining a good run. Plan is to train an ANN to give out expected lengths lost based on distance, going and bsp. I have now done first version of this with All Weather in mind and just by putting in criteria that run where actual lengths lost was less than half of expected distance behind winner was a good run and run where actual was more than 1.5 times the expected would be considered a poor run I were able to divide runs to roughly 25% good runs, 50% of ok runs and 25% of poor runs.

Next up is to check if this classification has any bearing on how horse performs next time out.

Trainer angle from the archives

I was looking for something else from my archives when I laid my eyes on some freebie system called The Marathon Chase System. As we are heading to jump season proper I decided to take a look inside as I didn’t recall how this particular one worked.

Well, inside was pretty simple system with a name of a trainer and criteria for distance the race is to be ran over as well as odds range for when to take the bet. System was available from, but I was not able to locate it from there anymore but to be on the safe side I am not going to give out the details of the system.

But I did check on the results in 2014 so far. Meagre 6 point profit over 36 bets so far and strike rate of 17%. Both 2012 and 2013 were profitable for this trainer and I had a look if there were other trainers who had been successfull with same criteria for races and there were. I was able to narrow it down to three trainers who had had sufficient amount of runners since 2012, wins were not only from super high SP’s and had consistent strikerate.

These three trainers have made a profit of over 150 points in last couple of years and up 20 points this year already. Which means that while, this is my first look at creating a trainer based system myself, I am going to elevate this to tracking and see how these perform for me.

Which is more important, strikerate or ROI?

I was looking through selections calculated by my bayesian system and came up with two approaches to a set of selections. Other is high strikerate low ROI and other is other way around. Naturally in perfect world one would take both, but what would you choose if you had to choose either or from below?

System A

  • 867 selections
  • 246 winners
  • Profit of 53.25 points
  • Strikerate of 28.37%
  • ROI of 6.14%

System B

  • 289 selections
  • 8 winners
  • Profit of 225.83 points
  • Strikerate of 2.77%
  • ROI of 78.14%

What would you choose?

1 2 3 4