Re: on ROI
 

[ QUOTE ]
3% ROI is on the high end IMO.

[/ QUOTE ]

Yeah, you're very wrong on this and so is your book. You can bet line movements for this and higher.

This is for normal sides/ML/totals etc. Your futures etc can and should push >50% if not higher before you start hedging etc.

Re: on ROI
 

[ QUOTE ]
Without a time frame, I'm not sure how to evaluate 2-3% ROI. For a week, it's obviously high. For a year, it's very low.

[/ QUOTE ]

ROI is clearly per dollar bet. Not for a time period.

Re: on ROI
 

how many bets is considered significant? What is the equivalent of say 100k hands of poker?

Re: on ROI
 

[ QUOTE ]
Yeah, you're very wrong on this and so is your book. You can bet line movements for this and higher.



[/ QUOTE ]

Please give me an example.

Re: on ROI
 

[ QUOTE ]
[ QUOTE ]
Yeah, you're very wrong on this and so is your book. You can bet line movements for this and higher.



[/ QUOTE ]

Please give me an example.

[/ QUOTE ]

Umm... Work back through the Pinny expected loss per dollar for the avg high volume arbitrageur. The long run is fairly quick with this. Avg negative ROI there is around 3.6 to 3.9 which implies at worst a corresponding positive ROI of at least this level at other books. If you throw in any actual bonus dollars (which I'm sure you are excluding and as I am for pure ROI discussion) you can quickly earn a fairly nice living. The swings are much greater, but I'm sure you can manage to get several nice plays per day at a decent Kelly stake with a mid five figure to low six figure bankroll and a small amount of talent.

There is really no reason to handicap games if you're going to earn such a small ROI, there are many people who are able to beat openers for double digit ROIs. There is really little reason to spend hours and hours analyzing information for an edge smaller than you can get just staring at a monitor.

Re: on ROI
 

[ QUOTE ]
how many bets is considered significant? What is the equivalent of say 100k hands of poker?

[/ QUOTE ]

It depends...

First of all, you don't really need to play 100K hands of poker to know if you are a winning poker player.

Let's say you have 50K hands in a certain online NLHE game (same site, same stakes) and you have a 14 PTBB (28 big blinds) win rate. The question is not if you are a winning player, but what your true winrate is. Many people on 2+2 don't seem to understand this distinction, which is why you see people making all kinds of absurd claims as to how many hands you have to play in order to "prove" you are a winning player.

It is the same with sportsbetting. It depends what kind of winrate you are putting up.

If you've placed 5K straight bets against a dimeline, and won 2900 of them (a incredible win pct of 58%), you are a winning bettor. The chance that you are just getting lucky is infintesmal.

If you've won 2750 of them (a very good win pct of 55%), the chances you are just running good is just 0.012%.

If you've won 2700 of these bets (win pct of 54%), the chances you are just running good is up to 1.2%.

In these examples, running good means that your true winrate is somehow known as being 52.4% (which would be breakeven on a dimeline).

If you wanted to know if you are bettor than a 50/50 handicapper (as opposed to wanting to know if you can beat a dimeline), then you would need much less bets in your sample. For example, if you win 1100 out of 2000 bets, you are definitely better than a 50/50 picker.

Most 2+2 nits seem to think you need to play 1M hands of poker (regardless of winrate), win 5800 out of 10K bets, or run a MLB Monte Carlo simulation 1M times to get a significant sample size. None of that is close to being true.

Ignore most of what you see on here with regards to sample sizes, unless you don't see the humor in the following exchange (from Dumb and Dumber):

Lloyd Christmas: What are the chances of you and me getting together?

Woman: 1 in a million.

Lloyd: So you're saying there's a chance!

Re: on ROI
 

I agree that 2% is pretty much the absolute minimum acceptable.
Don't believe me? Go here, plug in -100 and 51%.
3% seems reasonable, neither pathetically low nor absurdly high.

Re: on ROI
 

PO,

That was excellent. In the same token if you're winning at <1PTBB in a high variance style, you might need 5 million plus hands to determine if you are even a winner.

Re: on ROI
 

[ QUOTE ]
There is really no reason to handicap games if you're going to earn such a small ROI, there are many people who are able to beat openers for double digit ROIs. There is really little reason to spend hours and hours analyzing information for an edge smaller than you can get just staring at a monitor.

[/ QUOTE ]If my math is right, that means hitting over 57.5%, equivalent to a no-vig line of -136 (while getting -110). Is this seriously possible by just following line movement? Don't slow-moving books usually wait on posting their lines until they've stabilized a bit?

Edit: Also, are we talking daily sports (baseball/basketball) or football?

Re: on ROI
 

there goes thremp giving away his ricebowl again... somewhere on planet earth hogua is shedding a tear