Just Nine Lives

Just Nine Lives

If you read my previous article, you will know that the pursuit for data driven betting excellence continued last weekend and with a number of shock Premier League results now in the history books, it is probably best to not talk about last weekend.

A LOOK TO THE FUTURE

Before we press ahead with this weekend’s data driven football predictions and an attempt at gut instinct, lets take a quick look at the “league table” in terms of current performance:

Pure Gut Instinct: 50/98 (51.0%)
PremBot 1: 14/30 (46.6%)
PremBot 2: 9/20 (45.0%)
Hybrid Model (Combining First 68 matches and then PremBot only): 51/98 (52.0%)

As you can see the gut instinct is as about as average as it can be and it lead to an interesting question from the Twittersphere:

“Just what is the sample size that you will look for to start to make any real conclusions” or words to that effect

The answer to which is 100 and with us all but reaching that point in the Premier League season, we can draw some conclusions from gut instinct and quite simply, my own predictions skills are nothing more than ok.

Of course, there are some mitigating circumstances in all of this and that is the fact that away wins have been so prevalent, this season and when we look at the first 98 fixtures of the season, the outcomes have been as follows:

HW	36	36.73%
AW	43	43.88%
DRAW	19	19.39%
TOTAL	98	100.00%

Something which has the obvious ability to throw anybody in terms of what they think in terms of predictions and if these results have told us anything, it is trying to think counter-intuitive.

However, preceived bias and the weight of the “Big Six” is always something that is going to make that a lot more difficult and therefore, it has left me with a solid but not spectacular batting average in terms of success.

However, undeterred by such a poor performance last weekend, here are my own picks for the weekend:

Everton
Manchester City
Manchester United
Chelsea
West Brom/Crystal Palace Draw
Leicester
Tottenham
Liverpool
Brighton/Southampton Draw

This nine-fold is 402/1. A figure that is a lot less of a shot for the moon than previous week – not only becasuse we are a leg short but also because the predictions seem to be more in tune with the bookmakers.

Which also means, although we can make a conclusion when looking at the performance of gut instinct, it is far too early to make any calls regarding the performance of “PremBot” and his younger brother “PremBot Jr”.

Which leads us to where we are now and in a slightly curtailed Premier League, this is what the current form guide looks like:

If we used just the form guides as an indicator of predictions, here is what the nine-fold would be for the week:

Burnley
Manchester City
Manchester United
Chelsea
Crystal Palace
Leicester
Tottenham
Liverpool
Southampton

A no draw nine-fold which would pay out 579/1 for your efforts of being able to work out a higher number versus a lower number. However, we are going to look at what “PremBot” has served up and this is what he is plumping for this week:

A little more against the grain, and because of this the odds have gone out to 1,748/1. Something that seems slightly more achievable than the five figure payouts that were on the horizon in the past.

Of course, we cannot forget “PremBot Jr” and lets see what his nine-fold looks like by comparison:

Once again there is very little variance between the two (which makes me think, they might be too close too each other in the first place – although whether this is a bad thing or not can be argued) and with Crystal Palace tipped to beat West Brom instead of a Draw, the odds are shortened to 1,480/1

Another to look at, is just how the predictions fare against the bookmakers pre-match odds, a list that looks as follows:

Tomorrow, 05 Dec			1	X	2	B’s
12:30	Burnley – Everton		309/100	63/25	97/100	14
15:00	Manchester City – Fulham		3/25	91/10	201/10	14
17:30	West Ham – Manchester Utd		133/50	129/50	107/100	14
20:00	Chelsea – Leeds		27/50	71/20	497/100	14
06 Dec 2020			1	X	2	B’s
12:00	West Brom – Crystal Palace		191/100	56/25	159/100	14
14:15	Sheffield Utd – Leicester		289/100	63/25	101/100	14
16:30	Tottenham – Arsenal		103/100	64/25	283/100	14
19:15	Liverpool – Wolves		27/50	13/4	563/100	14
07 Dec 2020			1	X	2	B’s
20:00	Brighton – Southampton		77/50	12/5	187/100	14

The sticking points here are the draws at both the London Stadium and Anfield, that the data driven football predictions are suggesting. So this will be the biggest swing, although one must also remember that the bookmakers will not necessarily opt for draws themselves.

While they’ve also plumped for a Brighton win, while my models have suggested Southampton will. To be honest, we want to see variance such as this, because if those in charge of setting the prices cannot get it correct, all the time then it certainly gives me hope at the same time.

Last week, really was a week to forget on all fronts. But its too early to make any conclusions as to abject failure and this means, we are going to stick with the process for the forseeable. Wish me luck.

Happy punting and thanks for reading – to discuss this or any other projects you have, feel free to message me at either @dantracey1983 on Twitter or email me: dan@realfootballman.com

(THESE ARE NOT TIPS PER SE, THESE ARE JUST DATA DRIVEN FOOTBALL PREDICTIONS IN A PURE TEST AND LEARN MODELLING CAPACITY)