MLB Data Analysis

[durito]

Try to keep them straight. Mathy understands theory, and wins on his own merits. I can't remember MonkeyFocker ever making a serious contribution to HTT. He wants to play with the big boys, but hasn't done the work yet. If he is a winning player, he has shown no evidence of that here.

Clueless as always. You should read better.

How about we get someone like say ganch to set up a sports betting test regarding math/stats/modeling. You vs MF. I´m guessing i could get a few hundred k if not more backing MF.

[Justin7]

Clueless as always. You should read better.

How about we get someone like say ganch to set up a sports betting test regarding math/stats/modeling. You vs MF. I´m guessing i could get a few hundred k if not more backing MF.

I don't know him personally... Just from his posts. If he's a closet genius, he hides it well.

[bztips]

If you don't understand what Dark Horse said, you need to reconsider how you are doing your modeling. If scoring goes up 5% a year for 2 years, you're screwed if your models don't pick up on that -- even if it is stat driven, you'll get horrific results when both offense and defense are in alignment to make a higher or lower scoring game. Even in the course of a single season, the smaller changes can make a winning model useless.

I guess you and Dark Horse believe that most "math" guys estimate a statistical model once, and then blindly apply it, unchanged, year after year; never bothering to update the coefficients based on new data, and never testing new model structures.

You would be wrong.

[Justin7]

I guess you and Dark Horse believe that most "math" guys estimate a statistical model once, and then blindly apply it, unchanged, year after year; never bothering to update the coefficients based on new data, and never testing new model structures.

You would be wrong.

Didn't I say just the opposite?

[bztips]

Fair enough, let me restate. According to DH:

“Smaller sample sizes, that are just as real, but won't last from season to season. These shorter term fluctuations are available to those 'surfers' who know how to read the waves.”

“The underlying idea is that each season is different and that specific contours or outlines of the season will become clearer as the season progresses… In general the 'waves' appear to be more reliable towards the later part of the seasons.”

To me, it sounds like he’s saying that if you’re really good you can somehow identify these short-term ‘waves’ (also known as ‘normal statistical variation’ to us math sect guys) and use them to your betting advantage.

If you believe you can do that, with or without modeling, ok good luck to you. In that light, your example about what are in effect non-transient changes (eg., scoring rising by 5% a year) isn’t what DH appears to be hawking.

[MonkeyF0cker]

Try to keep them straight. Mathy understands theory, and wins on his own merits. I can't remember MonkeyFocker ever making a serious contribution to HTT. He wants to play with the big boys, but hasn't done the work yet. If he is a winning player, he has shown no evidence of that here.

LMAO. The moderator of the forum lowering himself to character assault. I guess I struck a nerve.

If you don't understand what Dark Horse said, you need to reconsider how you are doing your modeling. If scoring goes up 5% a year for 2 years, you're screwed if your models don't pick up on that -- even if it is stat driven, you'll get horrific results when both offense and defense are in alignment to make a higher or lower scoring game. Even in the course of a single season, the smaller changes can make a winning model useless.

OMG. Priceless! Yes. All of my data is from before two years ago. LMAO.

Where the hell do you come up with this crap? You and Dark Horse should write a book together.

[MonkeyF0cker]

HFA in arena is worth 1 point, not 3...At least it was before the restructuring of the league. I haven't calculated it since then.

LOL.

Unreal.

[durito]

I don't know him personally... Just from his posts. If he's a closet genius, he hides it well.

If you cant tell from his posts that he knows what he is doing then you are more clueless than i thougjt. Of course you wont back your opinion with money the same way you wouldnt bet thremp re sportsbetting/jazette.

Buy any models from wrecktangle recently? You do any of your own work?

Ill send mf a copy of your book, should be a good laugh.

[chunk]

Note to self: Never forget boxing gloves when going to HTT.

[Justin7]

Fair enough, let me restate. According to DH:

“Smaller sample sizes, that are just as real, but won't last from season to season. These shorter term fluctuations are available to those 'surfers' who know how to read the waves.”

“The underlying idea is that each season is different and that specific contours or outlines of the season will become clearer as the season progresses… In general the 'waves' appear to be more reliable towards the later part of the seasons.”

To me, it sounds like he’s saying that if you’re really good you can somehow identify these short-term ‘waves’ (also known as ‘normal statistical variation’ to us math sect guys) and use them to your betting advantage.

If you believe you can do that, with or without modeling, ok good luck to you. In that light, your example about what are in effect non-transient changes (eg., scoring rising by 5% a year) isn’t what DH appears to be hawking.

To me, he is saying that constants change from year to year. You should look for these changes earlier (instead of waiting for the season to end, and saying "Whoops! I was wrong assuming last season's observations would carry forward")

And the "clarity" of the changes becoming more apparent by season end... I think most people would agree that it is easier to cap a sport after 3/4 of the season, than 1/8th into it. Whatever changes there are much more apparent.

[MonkeyF0cker]

To me, he is saying that constants change

That pretty much says it all.

[evo34]

To me, he is saying that constants change from year to year. You should look for these changes earlier (instead of waiting for the season to end, and saying "Whoops! I was wrong assuming last season's observations would carry forward") And the "clarity" of the changes becoming more apparent by season end... I think most people would agree that it is easier to cap a sport after 3/4 of the season, than 1/8th into it. Whatever changes there are much more apparent.

Really? I would tend to think most sharp bettors make a better return over the first 75% of a season, vs. the last 25%. If you need 75% of the current season to adapt your model, your model is of no use.

[sideloaded]

daringly it's time to bring out the infraction guns. Don't let these fools style on you.

[Wrecktangle]

Try to keep them straight. Mathy understands theory, and wins on his own merits. I can't remember MonkeyFocker ever making a serious contribution to HTT. He wants to play with the big boys, but hasn't done the work yet. If he is a winning player, he has shown no evidence of that here.

Perhaps you're right, but if Mathy has his "theory" down, it is seriously flawed as he has eliminated whole "theoretical" areas for consideration. And, I see them both as temperamental bomb throwers.

[mathdotcom]

It still mystifies me how these debates just... suck.

Sports betting is so easy. All you need is intro stats and basic knowledge of a sport. Sometimes you barely even need the latter. Get relevant data and translate it into a prediction (i.e. odds). Compare your odds to the bookmaker's. If there is a discrepancy then pump and pound.

Some guys on here think it's impossible to do this because they want to know what a starting pitcher had for breakfast (oatmeal or McDonalds), and since they don't know they think there is no point trying to make their own odds because they don't have all the possible information. These guys are losers and should honestly be banned from a forum called "Think Tank". (I'm talking about Wrecktangle types, who are obviously unemployed welfare-collecting university drop outs that have an axe to grind with anyone who has knowledge). We have all disagreed with Justin/MF/durito/etc at times but at least we know what they're saying and why they're arguing about it.

[Dark Horse]

Really? I would tend to think most sharp bettors make a better return over the first 75% of a season, vs. the last 25%. If you need 75% of the current season to adapt your model, your model is of no use.

You're so obviously a beginner, talking theory without experience against guys who've been there and done that.

The first weeks of a season can be very easy. What could be the great mystery behind that?

[sideloaded]

You're so obviously a beginner, talking theory without experience against guys who've been there and done that.

The first weeks of a season can be very easy. What could be the great mystery behind that?

why is that? Does your model not take into account injuries or something?

[evo34]

You're so obviously a beginner, talking theory without experience against guys who've been there and done that. The first weeks of a season can be very easy. What could be the great mystery behind that?

Clearly I am a beginner. Yet somehow I realized about 15 years ago that turnover rates in the NFL are a contrary indicator (i.e., they are mostly luck and you want to fade abnormal past turnover rates). Contrast that with your excited post in 2010 about the fact that turnovers and points are correlated in NFL. Um, WTF? What exactly have you done; and where exactly have you been? I'm genuinely curious. Every shred of evidence points to you being a net loser. Would love to hear the counter-argument.

[bztips]

To me, he is saying that constants change from year to year. You should look for these changes earlier (instead of waiting for the season to end, and saying "Whoops! I was wrong assuming last season's observations would carry forward")

Frankly, I don't how anyone can be confident in interpreting what Dark Horse says. If he really meant that "constants change from year to year" he could have said that. Instead, he talks about waves and contours, and his other posts are equally full of mumbo jumbo. You literally have to try to interpret what the meaning of "is" is (thanks to Bill Clinton).

[Wrecktangle]

What the chest-pounders on this forum miss because of the size of their heads is the fact that you can learn things even from the newbies. Angles that "work" are based on some stat tendency and can usually be found if you look for it.

I don't begrudge anyone who can do well in handicapping. I even agree with Mathy about dumb-shit math being able to uncover +EV situations. I do think his attitude sux, but his drunken father probably beat him as a child (two years ago) when his Dad found out his mother was sleeping with the FEDEX man.