Team Contest question

[pavyracer]

Who was the genius that came up with the idea of unlimited bets?

[Nicky Santoro]

close to 100 games on the board today including college football.. and so far,i only have 5 plays.....

but some will start to take advantage at the end when they trail by 48 units and then start playing every side and total on the board to try to catch up, which will not help them at all.. not one bit..

[Mudcat]

Who was the genius that came up with the idea of unlimited bets?

Really? That's the rule? The SBR people have probably created some headaches for themselves with that.

[CaneDawg]

its fun

[pavyracer]

They can't be best bets if someone is allowed to bet them all every day. It's like a crap shoot Nicky.

[Tsoprano]

its fun

[Tsoprano]

There are plenty of contests/events (sbr best bet thread, 128 man contest, 256 man contest, parlay of the day, beat the prick, etc) where people can make their best bets, this is a fun shootout!

[pavyracer]

Here is the stupid rule:

Each of you must make a minimum 20 picks from now until December 30th. If any of you fall short of this #, a 1.50 unit penalty will be enforced for your team for each missing pick. No max on total plays.

Then rename the contest to the best shootout in contest history. This is definitely not a "Capper" contest. Name one capper that plays unlimited plays daily. Only squares do that.

[CaneDawg]

bet a bunch

either win a bunch

or lose a bunch

[pavyracer]

If the parameters for all the players are not the same then you can't tell who is the best. Is a pinch hitter hitting 0.500 on 100 at bats a better hitter than a 0.400 hitter on 1000 at bats?

[smitch124]

My betting finger is sprained...

[tacomax]

So the winner of the contest is the team up by the most units?

[englishmike]

So the winner of the contest is the team up by the most units?

Taco....this was a fantastic oppurtunity for sarcasm, I was gonna really fukkin drive some witty sarcasm your way.

Then I remembered you're an expert at thread forensics.

So yes, most units wins sir.

[tacomax]

So yes, most units wins sir.

Thanks, son - I couldn't locate the rules.

Competition Format + Non limited # of bets + Stale Lines = 1 Way To Play The Contest

[pavyracer]

Thanks, son - I couldn't locate the rules.

Competition Format + Non limited # of bets + Stale Lines = 1 Way To Play The Contest

Give this man all the winnings please.

[Willie Bee]

Why can't two people from the same household play? Hey, how come you're letting two people from the same household play? Why is there a maximum number of picks? Hey, how come there isn't a maximum number of pics? What, no moneylines? Why are there moneylines? Why does SBR discriminate and not have soccer? Hey, how come y'all have such a boring sport as soccer? Why are the lines going up so late? Why don't y'all let us play the later lines? Does that about cover it?

[englishmike]

Why can't two people from the same household play? Hey, how come you're letting two people from the same household play? Why is there a maximum number of picks? Hey, how come there isn't a maximum number of pics? What, no moneylines? Why are there moneylines? Why does SBR discriminate and not have soccer? Hey, how come y'all have such a boring sport as soccer? Why are the lines going up so late? Why don't y'all let us play the later lines? Does that about cover it?

I don't recall crying when i asked Lou if soccer was included but that wouldn't make for such a good story I suppose.

[pavyracer]

Hockey lines too...

[pimike]

Why can't two people from the same household play? Hey, how come you're letting two people from the same household play? Why is there a maximum number of picks? Hey, how come there isn't a maximum number of pics? What, no moneylines? Why are there moneylines? Why does SBR discriminate and not have soccer? Hey, how come y'all have such a boring sport as soccer? Why are the lines going up so late? Why don't y'all let us play the later lines? Does that about cover it?

Damn its a free tourney WTF SBR provides so much goodies for us, Yet I have never seen so many cry babies in my life!


Nice post SIR, you are right on

[daggerkobe]

Wow i finally agree with the forum Nazi.

[Ganchrow]

If the parameters for all the players are not the same then you can't tell who is the best. Is a pinch hitter hitting 0.500 on 100 at bats a better hitter than a 0.400 hitter on 1000 at bats?

In the example you've given, while we couldn't say for certain who were the better hitter, what we could say would be that there'd be a roughly 97.1% probability that the 0.500 hitter were a better hitter than his 0.400 hitting compatriot.

1. This obviously assumes both player's hitting abilities to be stationary processes.
2. This is only an approximation based on the t-distribution. The actual probability is closer to 97.4%.

[bbyhill]

There are plenty of contests/events (sbr best bet thread, 128 man contest, 256 man contest, parlay of the day, beat the prick, etc) where people can make their best bets, this is a fun shootout!

yes it is T

[pavyracer]

Ganchrow,

Then how come A-Rod gets paid 20 million a year and pinch hitters get paid 300K? I'm glad you work for SBR because if you were a MLB GM you would build a team worse than the Royals.

[Ganchrow]

Then how come A-Rod gets paid 20 million a year and pinch hitters get paid 300K? I'm glad you work for SBR because if you were a MLB GM you would build a team worse than the Royals.

You touch on an important point which does represent a certain flaw in my earlier analysis.

To put it in layman's terms, we know from our knowledge of MLB baseball that the probability of any MLB player being a "true" 0.500 hitter is vanishingly small. As such we could feel quite confident that our 0.500 pinch hitter simply managed to get extremely lucky during his 100 ABs. (A similar albeit less stark assessment could be made regarding our 0.400 hitter as well.)

The real problem is that I wasn't adjusting the realized batting averages for what we know we know about the realities of the game.

Now this doesn't mean that such a comparison couldn't readily be performed, just that there should have been an additional step that I omitted in the name of simplicity. My apologies.

To put it in more technical terms, the above analysis considered posterior evidence (0.400 over 1,000 AB and 0.500 over 100 AB) without regard to a prior distribution (or at best, implicitly assuming prior distributions to be uniformly distributed between 0.000 and 1.000 -- clearly erroneous.)

To more accurately perform this analysis, we'd need to consider the historical record and impute a probability distribution for "true" batting averages for MLB players selected at random.

The previous methodology utilizing a simple t-test would be emblematic of the "frequentist" school of thought, while the one outlined in this post would be strictly "Bayesian" in character.

The main reason why this was so relevant in the example given is because we were dealing with such highly improbable averages. If instead of averages of 0.500 and 0.400 we'd been given, say, 0.313 and 0.250, then we'd be able to feel much more secure in trusting the conclusions of the t-test without regard to our priors.

Nevertheless, the central thesis of my earlier post still stands. Just because the sample sizes of two performance records are unequal doesn't mean we can't draw meaningful probabilistic conclusions regarding relative skill levels.

[TPowell]

if anybody hasn't noticed, its more of a "who can chase steam the best" game. Frustrating to have lines that are 2-3 worse than the lines you have $ on