Tuesday, 29 May 2012

Cat Skinning

I was reading a self-review of the Football Elite service, and the opening paragraph in a section titled Over / Under 2.5 Goals was this:
I’m not generally a fan of backing things that are an unintended side consequence of something else but as the years go on it’s becoming more and more apparent that there is value in the unders with the sort of games the recommended bets tend to be.
On the face of it, such a correlation may seem surprising to many, but upon further analysis, perhaps not so much. That there is a correlation between my own XX Draws and the Unders markets is to be expected, because, as has been debated ad nauseum, the fewer goals there are, the higher the probability of a draw.

Matt's Recommended Bets are subjective, and typically his modus operandi (a lot of Latin in this post) is to seek underrated home teams playing more illustrious opponents. You can see the type of scenarios that Matt is looking for here. Matt finds teams in the same leagues that I follow, (in fact. my choice of leagues was influenced by Matt), whose strong home records have been under appreciated by the market.

Last season, several teams featured multiple times, e.g. Hannover '96 were selected 11 times, with a record of 7 wins and  4 draws. Hannover alone contributed 8.25 points of the season's profits. Other teams chosen multiple times were Ajaccio (selected 8 times, 3 wins, 4 draws, 1 loss), Atalanta (6, W3, D3) and Evian Thonon Gaillard (6, W4, D1, L1). Those last three added another 6.85 points.

It should be noted that Football Elite's Recommended Bets are not always to win straight up - often the selections is a Draw No Bet, +0.5 or +0.25, and these bets alone showed a profit of 11.63 points on the season.

Overall, Matt's record from 96 match selections was 37 - 34 - 25, for a profit of  7.81 points. Overall, Football Elite's profit on the season was 14.75 points when the ante-post (more Latin for you) bets are added in, but since they are a different type of bet altogether, and I think it confuses the situation to lump these in with the match selections.

Anyway, 7.81 points from 96 selections, an ROI of 8.1%, is decent enough, and indeed, as Matt notes, there is also an upside to backing the Unders in these games. This would have generated another 5.76 points of profit, not insignificant.

It's interesting that a strategy of simply backing Hannover '96 those eleven times, and doing nothing else, would have generated more profit, and an ROI of 75%! But hindsight is wonderful.

What I find most interesting perhaps, is that the character of matches that Matt ends up selecting, and the matches the XX Draw spreadsheet selects, are quite similar. In broad terms, the spreadsheet identifies matches where a combination of recent form and long-term rating, when adjusted for home advantage, predicts a match likely to be low scoring. Matt's selection process identifies matches where the home team appears to be value to beat the away team, and because Matt seldom looks at odds-on selections, the matches he ends up with are games with no clear favourite, and thus the draw is on the shorter side than average. Running the numbers, the average draw price on FE's selections last season was 3.43. This may seem a little short, but the really interesting thing is that if the draw had been backed in those 96 selections, rather than the home team, the profits would have been 19.64 points on the season.

That's 19.64 points on the draw versus 7.81 points on the home team, an increase in ROI to 20.5 %. And these numbers are from using the data at football-data.co.uk. It wouldn't be unreasonable to expect that with bets in early on the exchanges, those prices could often be beaten.

The FE selections were under 2.5 goals 58% of the time, and the average goals per game was 2.42. The fact that the edge on the unders was fairly modest means that the observations above could be a blip, but it would be interesting to hear from Matt how his match selections would have fared in previous seasons if backed as draws.

Most football services that I am aware of, admittedly not many, seem to go for selections in the 2.0 to 3.0 range, thus avoiding both odds-on selections and the draw. Understandably most services, or more likely their clients, want a steady flow of winners, at better than evens (if only slightly) but I think this is a mistake. If the numbers from previous seasons are even close to 2011-12, Matt, and his clients, would be far better served by targeting the draw, albeit with its inevitable longer losing sequences. The evidence so far shows that this is where the value is.

None of the above observations are in any meant to be a criticism of Football Elite, who have been out there for a while with many happy customers. In the same way that I didn't immediately connect the dots and realise the potential of Unders betting along with the draw for my XX selections, I'm simply pointing out that there may be other, more profitable, ways to play selections, if the risk tolerance of clients is high enough of course.

I have seen a few comments about the correlation between Under / Over prices, and the TOFG (Time of First Goal) markets, and someone (I forget who, sorry) recently left a comment suggesting that the Half-Time 0-0 market might be another opportunity. It certainly should be, and is something I'll be tracking next season. For the record, in 2011-12, of the 156 XX Draw selections, 60 were 0-0 at half-time (38.5%, IP 2.6) and the average TOFG (when there was one) was 33.66 minutes. Unfortunately, I do not know what the HT 0-0 prices were, nor what the average TOFG is, but I would hope that both offer some value.


Maurice said...

Let's hope it will be another good year with the XX draws.
The 0-0 Half-time was my idea. It's not a big leap from unders to 0-0, to 0-0 HT, so maybe others have thought about that too. I also want to test that coming season with low stakes. I will use the classic XX draws and not the extended.
Greetings, Maurice

Matt FElite said...

Hi Cassini,

This was something I looked into when I was doing the review, primarily because as you say I noticed there was a fair bit of crossover with the FE bets and your XX bets and they are typically the same "type" of games.

Unfortunately I think it was just a random case of the recommended bets happening to hit a lot of draws this season (34.3%). I wouldn't expect that long term.

The shortlist bets this season only had 28 draws from 117 bets (23.9%)

2010/11 season there were 22 draws from 97 bets (22.6%) for the recommended bets and 39 draws from 152 bets (25.6%) for the shortlist bets.

2009/10 season before that it was 31 draws from 121 bets (25.6%) for the recommended bets and 50 draws from 129 bets for the shortlist (38.7%).

2008/09 season the rec bets and shortlist bets were all together as one and the draw rate was around the 25% mark from memory.

So I dont think there's much mileage in backing the draw blindly although there certainly could be an improvement with manual intervention.

I think it is interesting that we both pretty much target the same kind of games but bet on different outcomes, yet both make a long term profit. Before the start of the season we will have to compare results and have a look at what happens when there is a crossover of FE bets and XX bets. Could be an interesting angle there.