SportPunter.com - Predictions and Profitable gambling in Sport

Welcome to the Sportpunter Major League Baseball Report Card. Here we will look closely at how we went in 2005 and compare this to previous years and well as forecast how we think we will go in 2006 and beyond. We will also look at in depth data in order to maximise your profits using the proper money management. Please feel free to comment on the analysis in the forum thread shown here:

The MLB Baseball Season in 2005 was a huge success. We started the year giving head to head probabilities and suggested bets, and then half way through the season we introduced the totals betting which outperformed anyone’s expectations. Including in the model mid-year we introduced weather and wind conditions into the model including temperature, sun/rain conditions, wind direction and speed as well as ground surface conditions.

The results were a huge %ROI for the year and big profits achieved by all. The results for 2005 as well as previous years are listed below:

Year	#Bets	#Wins	%Wins	$Bet	$Profit	%ROI
TOTAL	2054	956	46.5%	$312,313.27	$24,261.26	7.8%

2005	823	384	46.7%	$117,981.29	$8,039.23	6.8%
2004	823	393	48%	$138,378.31	$11,791.63	8.5%
2003	408	179	44%	$55,953.67	$4,430.40	7.9%

Year	#Bets	#Wins	%Wins	$Bet	$Profit	%ROI
TOTAL	4404	2196	50%	$458,456.69	$35,820.09	7.8%

2005	1431	721	50.4%	$152,691.34	$15,119.93	9.9%
2004	2048	1015	50%	$210,886.14	$14,978.33	7.1%
2003	925	460	50%	$94,879.21	$5,721.83	6.0%

As shown above, the head to head record was slightly under those of years 2005 and 2004, whilst the totals betting probably over performed and recorded a higher %ROI to the previous 2 years (despite a similar $Profit to 2004).

The graphs show that for head to head betting, we achieved large profits until about 40% of the way through the year, followed by a dramatic decrease in profits. Profits of the first section on the year were cut in half, but then this was followed by great profits once again to finish the year about the same stage as what we were 40% of the way through the year.

Totals betting was a lot more consistent, and showed steady increase throughout the year. At most a drop of around 15% was recorded mid-year, but this was soon won again.

It is believed, that the totals betting strongly over performed in 2005, and we anticipate a profit of anywhere around 3% to 7% ROI for 2006. We do not believe that the totals betting is in any way superior to the head to head betting, and one, should they be subscribing to the baseball, bet both options with equal weighting.

Shown below is the results for MLB given in three tables. The tables are for head to head betting only. The first table is for all results, the second table for only betting on underdogs (where odds at equal to or greater than 1.91) and the third for favourites (where odds are less than 1.91)

As shown above, we made a slight 0.8% ROI betting on all outcomes. Betting on outcomes with probabilities less than 60% was most profitable, likewise with odds of $1.90 and above. This can be better shown in the analysis of underdogs and favourites as shown below:

A very healthy 6.8% ROI was made betting on all overlays where the odds are 1.91 or greater. A 9.2% ROI profit was made betting on underdogs where the probability is greater then 50%. Obviously this implies that the greater the overlay, the greater the profit which is shown in the last section where we made 29% ROI for overlays 30% or greater, and 9.4% ROI for overlays of 12.5% or greater. As throughout the year we bet on 5% or greater overlays in 2005, a 7% ROI profit was made. Given this, subscribers might want to increase their minimum overlay to 12.5%, however lack of data suggests that keeping it at 5% would be very satisfactory.

In short, there was no room whatsoever to bet on favourites. It was hoped that a minimum overlay would occur where we might be able to gain on favourites that have large overlays. This isn’t the case. In 2005 favourites were avoided, and there is every reason to avoid them again.

In the analysis of totals betting. We have the following tables for all results, betting on overs and betting on unders respectively.

Strong profits recorded, moreso when just looking at when the probability is 50% or greater. Having a minimum overlay of 6% increased the return on investment to nearly 12%. More in depth analysis shown in the following tables.

About one third of the profits in totals is obtained from betting overs. Around 20% less amount bet as well. Betting when the probability is 52% or greater increases profit, whilst a minimum overlay of around 6% increases the profit to 11.1% ROI. This might be something that clients might want to consider when betting overs. However a relatively small sample size suggests that this might be looking too closely at the data post hoc.

Two thirds of the money made on totals come from unders betting, with at least 20% more money bet and 20% more bets made. The success rate of bets was 5% greater, which shows a possibly bias in the odds, favouring the punter in the under. More than half the money made betting unders was done so when the probability was 62% or greater. Obviously outlining when greater overlays occur, this is backed up by the fact that almost 80% of the money gained is when the overlay is greater than 22.5%. No minimum overlay is shown to be beneficial. All range of odds satisfied profits.

Sportpunter has also looked at the different odds/probs and overlays from all it's data going back to 2003. The results are shown in the following links:

When looking at the optimal Kelly Fractional we have to work out the expected profits given the overlays. Please note that any calculated optimal Kelly fractional is dependant on the previous years data and is by no reason a Kelly fractional that will increase profits the most in future years. It is purely a means of what would have been the best Kelly Fractional in the past, and quite possibly might be in the future.

Given the relationship described based on favourites and underdogs betting in section 2, we will only work out the expected profits based on betting on underdogs with odds of 1.91 or greater for head to head betting.

Expected profits in 2005 for head to head betting on favourites was $30,693.49. This is a %ROI of 21.8%. As the actual %ROI was 6.8% ROI, this equals therefore an optimal Kelly fractional of 21.8/6.8 = 3.2. Or in other words 1/3.2th Kelly fractional. Hence our suggested 1/4th Kelly fractional seems very good, as it is accurate yet conservative. Subscribers should not therefore use a Kelly fractional of 3 or less, and should use 3.2 or greater. The greater the Kelly fractional, the less the risk.

Expected profits in 2005 for totals betting was $23,271.49 which gives a %ROI of 15.8%. Given the actual %ROI was 10.4% this suggests an optimal Kelly fractional of 15.8/10.4 = 1.5. Or in other words 1/1.5th Kelly fractional. This is incredibly low and we have to discuss the meaning behind this.

The very low optimal Kelly fractional outlines the huge year that we had in 2005 betting on totals. The fact that we are the only model to have weather and wind conditions into the model, outlines it’s superiority. Not only do we have the weather conditions but also account for error in predicted weather and actual weather. It could well be that these added extras into the totals have made it so successful. Notice in section 2 that the overlays that we have got are lower than that in the head to head betting, which is either a sign of the accuracy of the bookmakers or the accuracy of the totals model. Given our profits, it’s definitely the accuracy of the totals model.

It could also be that the totals model has over performed in the last year, just due to luck. It is the opinion of us at Sportpunter that this might well have been the case. More of this to follow in section 5, but in the meantime I would suggest that clients should use at most 1/3rd Kelly. Whilst 1/2nd Kelly seems to be conservative, at least by last years results, it could also prove to be costly. We recommend that you use 1/3rd Kelly on totals at most, and more conservative gamblers might wish to use 1/4th Kelly fractional or greater.

Basically with a $10k bank the following are the results betting on head to heads as well as totals. One bank is used for both in the following analysis:

The constant profits as shown above in the constant Kelly bank obviously show the consistency in the model in 2005. The difference between the constant Kelly graph and the moving bank is very dramatic. The surreal bank of the moving bank is indeed unbelievably large, but this outlines the advantage of using a moving bank over a constant bank should you be betting with profits. The moving Kelly graph is based on an adjusted bank at the start of every week.

A great year for Baseball betting in 2005. And there is no reason why the success will not continue in 2006. Previous years suggest that 2005 was on relatively equal par with that of previous years. It is hypothesised that totals betting will not be as successful as it was in 2005 and that it over performed a little. That being said, we expect to get anywhere between 3% and 7% ROI for totals, which over a large amount of bets, is highly respectable. With head to head betting returning similar figures, 2006 looks to be a good year for gambling.

That being said, Sportpunter are always keen to improve the model. At the moment we are looking at umpire decisions on the game and their influence over totals. Whilst it is quite clear that some umpires let more home runs go freely than others, we are in the middle of determining if this is in fact significant and also if it is a good predictor variable.

Despite this, Sportpunter’s MLB model is clearly a class above the rest, having unbiased mathematically based predictions independent of media hype. The model maintains memory of all games, and establishes the importance of variables and their influence over the game; for example weather, and wind conditions.

We hope that if you enjoyed the fruits of Sportpunter’s baseball in 2005 that you would not hesitate to sign up again in 2006. And if you are new to Sportpunter’s baseball model, then we recommend that you stop looking at the profits we have made in the past and present (as shown above), and join in to also reap the rewards.