Wednesday 4 May 2011

I rest my case your honour......

After writing my latest post yesterday and reading through it again today at lunchtime, it did make me wonder.....What if by including the 2006/7 and 2007/8 seasons in my results, this has been causing me to think things were behaving differently from the past when in reality, they weren't that different from the last couple of seasons?

As most people know now, from when I started out on this journey, I've only been quoting the last 4 season's results as the data I used to build the ratings from 2002-2006 is all backfitted data. If I was quoting these results, then I would be painting a misleading picture as the results for these years had ROI's of about 40% as I fixed the parameters of the two rating algorithms to maximise these results.

However, as I've quoted a fair few times now, I also used a sample of data from the 2006/7 and 2007/8 seasons to ensure my model wasn't using data from too long ago to miss any changes in trends. I randomly selected 50% of the games from 2006/7 and 2007/8 to include in my backfitted model and therefore, these years will always be slightly overstated. However, these seasons account for 1/6 (1/3 * 50%) of my full backfitted results and therefore, I was still fairly confident that the results weren't overstated in these seasons.

Admittedly, it would have been much better to just ignore the first two seasons completely and only look at the full 2 seasons of backtested results but my issue there was I only had 2 season's of data and it meant my sample size was only 5,164 games. I noticed a post by Cassini today about someone saying you needed 100,000 bets to be sure of an edge, so I'd be paper trading for the next 20 seasons under this scenario!

Anyway, I thought 5,164 games was too small to draw meaningful conclusions and that's why I've always went with 4 season's. 4 season's gave me 12,037 games and therefore, I was happy to go with this, on the understanding that my first two years were probably slightly inflated by around 20% I guess (hard to know for sure tbh as it depends how good the backfitting was in these 2 years compared to the backtested results!)

After writing yesterday's post, I could see clearly that the draw % was much lower in the first two seasons and therefore, this would be bringing down my overall draw % average. Hence, maybe this season's wasn't as bad compared to the last two season's backtested results.........



There is a lot of information to glean from this but whatever way you look at this, it shows my worries were unfounded. The one thing that stands out here though is how consistent the 08/09 and 09/10 seasons were. They had a very similar H/A/X % across both years. The draw % was remarkably similar with 26.1% in the first season and 26.4% in the second season.

To put this into perspective, the draw % for this season is 31.3%. That's an increase of a full 5% points or an increase of around 18% on last season.

Over the course of this season (2801 games), we can use the same draw % as last season at 26.4% to get an expected number of draws this season. This would be 740 draws. What have we had this season? 876 draws. Hence, 146 draws more than you would have expected, based on last season.

How much does this hurt the systems? Well, for every draw, the systems score -1pt. The average odds for my all bets is around the 7/4 mark (2.75). Using this as a proxy, we can say that each draw costs the systems 2.75pts (1.75pt profit as against 1pt loss). Hence, the systems have lost 400pts this season compared to last season.

This season, the systems have ONLY made a profit of 203pts with an ROI of 7.3%. If we add on the missing 400pts (sounds a lot when I say it!), then we'd be looking at an ROI of 21% for this season.

What was the ROI in the 2009/10 season? 21%. What was the ROI in the 2008/09 season? 21%.

I rest my case your honour.

Incidentally, the attachment also shows the draws by each system for the last 3 seasons. You can see just how badly some systems have been crippled this season by the draw.....

System 7 has a 32% draw rate this season compared to circa. 25% for the last two seasons
System 8 has a 30% draw rate this season compared to circa. 25% for the last two seasons
System 22 has a 33% draw rate this season compared to circa. 28% for the last two seasons
System 23 has a 36% draw rate (50% after Christmas!) this season compared to circa. 24% last season
All systems 6-21,6-22 et. al have draw rates much larger than the last two seasons as a result of this too

Whatever way you look at the data, this draw effect isn't caused by me backfitting results, having poor ratings, building poor systems etc. It is just short-term variance that has unfortunately lasted for half a season. Long-term though, the work I've done on this project isn't flawed I hope (and pray!)

6 comments:

  1. hey,

    I am loving this blog, just spent the past hour reading through all these posts.

    Would you be interested in exchanging blogroll links?

    My site is http://sportstradinglife.com/

    Let me know if you want to add me and I will add you too!

    Keep up the good work!

    ReplyDelete
  2. Hi Ben.

    I've added a link and I'll take a look at your blog when I get some time mate.

    Cheers,

    Graeme

    ReplyDelete
  3. Your New Blog Reader8 May 2011 at 20:10

    Hi,

    I've checked the individual draws for the 5 major leagues in England and the SPL in Scotland. I came up with the following numbers:

    Premiership - 107 draws (29.89%) way higher than the last 2 seasons. 96 draws(25.3%) and 97 draws(25.5%) for 09-10 and 08-09 seasons.

    Championship - 148 draws (26.81%) performed better than the last 2 seasons. Both 162 draws(29.3%) for 09-10 and 08-09 seasons.

    League 1 - 137 draws (24.82%) performed equally with the 08-09 season. 152 draws(27.5%) for the 09-10 season.

    League 2 - 168 draws (30.43%) higher than the previous 2 seasons. 141 draws(25.5%) and 164 draws(29.7%) for 09-10 and 08-09 seasons.

    BSq Prem - 159 draws (28.7%) higher than the last 2 seasons. 155 draws(28%) and 142 draws(25.7%) for 09-10 and 08-09 seasons.

    SPL - 47 draws (21.76%) better than the last 2 season. 54 draws(27.3%) and 51 draws(25.8%) for 09-10 and 08-09 seasons.


    If I add up the draws for these 6 leagues I would get 766 draws for the 10-11 season.

    09-10 season: 760 draws
    08-09 season: 753 draws

    According to my computation, the draws for this season are indeed higher than the previous seasons but not high enough to be the cause of the system's problems.

    However, your post says that there are 876 draws for this season. Did I make a mistake somewhere? Maybe I added a league that should not be there or omitted a league that was being used? By the way, the results for this season come from futbol24. I hope you can clear this up for me.

    Cheers!

    ReplyDelete
  4. Hiya mate.

    Being honest, I hadn’t even looked at the draw % by league as it’s not something I track. I didn’t think there had been an increase in the number of draws in any of the leagues this season and I think your research backs this up!

    Ultimately, all my figures only look at the games where my ratings decide there is value to be had by backing one of the two teams. That’s the key for me as at the end of the day, I’m not selecting any games where there is an odds on team which is priced 1.5 or less and therefore, I miss out all the Man Utd/Arsenal/Chelsea/Rangers/Celtic games in the main. I also miss out any games where the home team is 3/1+, so that rules out many of these games that involve these teams away from home too. Obviously, I miss out some other games too involving teams at the top versus teams at the bottom that meets these odds.

    I don’t recall moaning once this season about the number of draws in any of the leagues. I can recall two times this where a league has found 3 draws in 12 games and my 3 bets were the draws! Hence, the draw % is 25% in the league and my draw % is 100%!

    All of the stats I quote are purely for the games involving teams where my ratings think there is value as that’s my sample size to pick the bets from. This weekend, there were 12 games involving teams my ratings thought were value and therefore, that’s the 12 games I’m looking at. I’m not really interested in the number of draws in other games as I don’t bet in these games.

    Obviously, when I build new ratings for my 3rd algorithm or adjust my ratings in the Summer, I’ll bring in the full fixture list as if I adjust my ratings, I’ll capture different games historically.

    With regards to the 876 draws (now 887 after this weekend after 11 draws from the 76 games), that is from the 2,877 games where my systems have selected a team to bet on. That isn’t the number of draws in any of the leagues etc.

    The only time these numbers become relevant is when you are looking at past seasons for the systems. In terms of any season, you can’t tell if this makes sense or not as it’s only comparable to the systems performance. Hence, this season has seen the ratings find a lot more draws than usual since Christmas. There hasn’t actually been a lot more draw in the league since Christmas, it’s just in games where my ratings decide a team is value.

    I hope this makes sense. I can’t recall saying that there are more draws this season in any of the leagues as it’s not something I track. If the draw % was running at what it has been running at since Christmas for my systems, the bookies would be shortening the odds on games to finish a draw going forward! Lol

    Apologies if I led you on a wild goose chase!

    Graeme

    ReplyDelete
  5. Your New Blog Reader9 May 2011 at 09:31

    Hi,

    Thanks for clearing it up. It seems I made a mistake in assuming that you were referring to the whole leagues. Sorry about that.

    I see that the systems are doing well for May. I hope it ends on a good note!

    Cheers!

    ReplyDelete
  6. No probs mate. Easy mistake to make!

    It's hard at times to keep up to speed with things as all my analysis in the first half of the season sits in the TFA forum as I wanted to keep the blog purely for results tracking. However, I wanted a bigger audience after Christmas and that's why I've started posting all this analysis on the blog and not just in the forum.

    Admittedly, anyone who hasn't been reading from day one might not have a clue what's going on but I try my best to do regular posts summarising how things have gone so far etc. I'll be spending a lot of time looking back on this season once we get to the end of the season, so people will be able to catch up then I hope.

    Yeah, it was good that May started a bit better than April finished. Even then, I sort of assumed that Newcastle would let an equaliser in as that's the sort of thing that has been happening all season but thankfully, they held on against 10 men! The margin at this game between success and failure is so thin and I really think I've been on the end of some shocking luck since Christmas with teams throwing away leads and only drawing but long-term, you just have to try to see past this sort of thing.

    Hopefully the season will end well and I can get building towards next season again. Whatever happens, I would hope to beat my returns from this season next season as I've learnt a lot and have lots of ideas around ways to improve the profitability of my ratings, so I'm sure I can do better next season. Time will tell though.

    Graeme

    ReplyDelete