[Pedro]

Hello There!

I was wondering if anyone has any idea of obtaining/calculating the alternate football spreads on pinnacle since they are NOT included in the XML data download.

I have requested an algorythm from Pinnacle but they were not willing to share it with me.

It would be great if someone would have an idea regarding the above.

Thank you,

Pedro

[bigboydan]

I have requested an algorythm from Pinnacle but they were not willing to share it with me.

i'm sure you knew there answer before you asked the question though.

[Pedro]

i'm sure you knew there answer before you asked the question though.

Well, that is usually their answer but I am thinking if they are willing to share their odds (with their XML) why wouldn't try to share their alternate lines in one way or another.

Pedro

[Ganchrow]

I was wondering if anyone has any idea of obtaining/calculating the alternate football spreads on pinnacle since they are NOT included in the XML data download.

Any algorithm to properly calculate this can not be determined from first principles but would need to be derived from a scoring distribution for specific spread/total combinations.

In practice however, by jointly considering spreads AND totals you'd likely lack sufficient time relevant data to produce statistically meaningful results. So as a simplification, you might choose to ignore totals and limit your domain to spreads or even perhaps just spread tranches (in other words, for a spread of 13.5 you may choose to look at scoring distribution of spreads from 12.5-14.5).

It is this latter tack which Stanford Wong follows in his book Sharp Sports Betting. Now I don't have the book in front of me, but if I'm not mistaken, Wong only derives with NFL spread equivalences for spreads of 3.5 points or less. Still SSB would nevertheless show you the process of coming up with your own spread equivalency chart.

Still, this might not be exactly for what you're looking. I have no idea how Pinnacle determines its own spread equivalencies, and although we can assume it's nearly identical to the method outlines by Wong, we still don't know what data set Pinnacle has chosen to consider, nor do we know if they may have chosen to transform the data based upon their lines makers' observations. Hence, even if you do come up with a statistically valid equivalency chart, there is really no guarantee it will match Pinnacle alternate spreads exactly and what’s more there's likely to be systematic bias in its predictions.

While creating your own equivalency chart is probably a good exercise in simple statistical modeling, if your goal is to nonjudgmentally determine Pinnacle's alternate spreads, you'd be better served following a different strategy.

If you have basic programming experience, you could easily write an HTML scraper script to extract any data you desire from Pinnacle (or any other sportsbook's web pages). From my experience, due to its facility with text parsing, it speed, and its simplicity, Perl is a perfect language for something like this. My only advice here would be to make certain you don't run this script too frequently as Pinnacle is always on the lookout for bandwidth drains and unless your very clever with manipulating HTTP headers, they'll know you hit their page with an automated process (of which they also do not approve). Even once a minute may be too much. Keeping it down to once every five minutes you should be just fine.

If programming isn't your bag (or if you just decide you'd rather determine alternate spreads ex ante), you may want to compile either a database or a spreadsheet of as much historic data relating spreads to alternate spreads. If you record data a few times a day Monday through Saturday (and considerably more frequently leading up to game times on Sunday), you should have a decent data set after two or three weeks. From that you should easily be able to interpolate most missing data points. The only problem here is that Pinnacle could subtly change their algorithm at any point without notice and you’d left with incorrect figures.

[raiders72001]

distribution for specific spread/total combinations.

Mistake that many people make. I was discussing the valueof 1/2 point with someone and he didn't understand how the value of a 1/2 point in the NFL is different at a total of 34 compared to a total of 51.

Pedro mentioned this one time before but instead of using a database and computing your own formula it's easier to just use Pinny or 5Dimes to compute the information that you are looking for.

Another way of doing it is to pay someone that knows nothing about gambling, but has a good math background to create algorithms for you. You supply him the info you want and he does the work. Many guys do this.

[raiders72001]

I really didn't learn much from Wong's SSB book except when he discussed poisson distribution. I think there are a lot better books at the gambling book store in Vegas.

[Ganchrow]

I really didn't learn much from Wong's SSB book except when he discussed poisson distribution. I think there are a lot better books at the gambling book store in Vegas.

Speaking of Poisson distributions did you notice that Wong claimed that in baseball total runs scored in a game follows a Poisson distribution? Clearly, that's not even true for a single team let alone both. That claim seemed to seriously belie Wong's knowledge of baseball.

Anyway there may well be better than SSB out there. Matter of opinion but I happen to think SSB is a great primer for those new to either stats or gambling.

[Ganchrow]

Mistake that many people make. I was discussing the value of 1/2 point with someone and he didn't understand how the value of a 1/2 point in the NFL is different at a total of 34 compared to a total of 51.

Perhaps you then need to find a new discussion partner. But seriously, I don't wager on football at all, nor do I follow either Pinnacle's or 5Dimes's alternate lines so I'm simply ignorant to how much either of those books in practice choose to allow the total to affect their half (or more) point prices. Quickly eyeballing Pinnacle, it seems the answer may well be "very little".

Another way of doing it is to pay someone that knows nothing about gambling, but has a good math background to create algorithms for you. You supply him the info you want and he does the work.

Yes, but doing it yourself is half the fun.

[raiders72001]

Yes, but doing it yourself is half the fun.

I agree with you about SSB being a nice primer. Honstly I don't remember everything in the book but I did look at all the charts and know that it was interesting also.

[raiders72001]

I don't wager on football at all

what sports do you wager on?

[Pedro]

Thanks for the suggestions.

I have been getting their odds off before but my ip got blocked so I was hoping to extrapolate it from their XML data.

I will probably go ahead with getting their odds off every 10 minutes or so and build a database of their data.

Thanks again,

Pedro

[Ganchrow]

I will probably go ahead with getting their odds off every 10 minutes or so ...

Lucky odds.

[Ganchrow]

what sports do you wager on?

Baseball and baketball (NBA and NCAAB),

[JoshW]

I really didn't learn much from Wong's SSB book except when he discussed poisson distribution. I think there are a lot better books at the gambling book store in Vegas.

Care to share any of these titles with us? SSB was the best sports betting book I have read, but I will admit to not reading too many. Most of the books are filled with so much misinformation and horrible writing, figured was best just to stay away.

[Ganchrow]

Care to share any of these titles with us? SSB was the best sports betting book I have read, but I will admit to not reading too many. Most of the books are filled with so much misinformation and horrible writing, figured was best just to stay away.

Fixed Odds Sports Betting: Statistical Forecasting and Risk Management by Jospeh Buchdahl is quite decent. It's written for a European audience so he does make some rather euro-centric "football" claims such as "bet singles" and "back favourites or short prices". Still, his reasoning and methodology are quite satisfactory. He spends several chapters going over the results of Monte Carlo simulations, which does get tedious, but his results are quite illuminating with regards to money management techniques. Other than that he doesn't get bogged down in details and the book seems shorter than its 225 pages. All in all, however, a decent read for novice gamblers with a rudimentary understanding of probability.

[JoshW]

Very cool ganchrow. I will be sure to check it out.