How Will you play this?

**sharpcat** · 02-13-10, 04:56 PM

No bet here

**xyz** · 02-13-10, 05:09 PM

Check the half point calculator and see the probability of game ending with a total of 129. Then compute your edge, if there is any.

**Dunder** · 02-13-10, 05:22 PM

This is another one of those Polish Middles that you seem to like.

You do have an edge if betting both sides, essentially trying to collect 3.44% against the 2.5 (or so) percent probability of the game ending on 129. It comes back to the same sort of issue as you asked before.

Applying Kelly will tell you that a large proportion (circa 20%) of you bankroll should be risked here. I would maintain that the figures involved are too marginal to do so, others will disagree.

If you don´t know which line offers the +EV opportunity then leave the game alone would be my advice.

**blix177** · 02-13-10, 05:24 PM

It is hard to predict which side has the +EV with 2 min before game starts. I barely have the time to put in the bet much less, do the math. And what I notice is I normally can't find these lines unless it is right before the game, when all the action is coming in.

**Dunder** · 02-13-10, 05:34 PM

Originally posted by blix177

It is hard to predict which side has the +EV with 2 min before game starts. I barely have the time to put in the bet much less, do the math. And what I notice is I normally can't find these lines unless it is right before the game, when all the action is coming in.

I understand that. In general terms most people who exploit differentials in lines these days use software that scrapes the books or compiles their XML feeds and immediately highlights lines that are off-market. As a rule the line that is off-market is the one most likely to be +EV.

Trying to do this sort of thing manually, is a lot or work and is often going to provide an incomplete picture.

Might be an avenue to explore.

**whatsgood5** · 02-13-10, 05:47 PM

I'd personally go C here

**Peep** · 02-14-10, 12:34 AM

I make it closer to 2% the number will fall exactly on the 129.

I have 242 games in the DB lined at 128.5, 129, and 129.5. Only 4 hit the 129 exactly as a closing score.

**u21c3f6** · 02-14-10, 10:46 AM

I would choose option C. I will assume Dunder's 2.5% fail rate. I also set both sides of a hedge to win the same amount. While setting both sides to win equal amounts is not exactly optimum, it fits with my personality and is much easier for me to handle.

Given the above criteria, if I wager 105 on the -105, then I need to wager 93.18 on the +120. This gives me 198.18 risk (2.5%) for 6.82 gain (97.5%). This gives me a Kelly of approx 25%. I personally use half-Kelly and would risk a total of 12.5% of my bankroll if I were to wager on this particular hedge.

I would have no problem using the above calculations but if the fail rate was really only 2% (and I don't particularly believe that), then Kelly would be approx 40% and even half-Kelly at 20% would be above my comfort zone.

Joe.

**tachi** · 02-14-10, 02:54 PM

Originally posted by Dunder

Applying Kelly will tell you that a large proportion (circa 20%) of you bankroll should be risked here.

Kelly,Melly,Selly..

no system can force me to bet 20%.

Dunder,how much times you bet 20%,honestly?

**u21c3f6** · 02-14-10, 06:06 PM

Originally posted by tachi

Kelly,Melly,Selly..

no system can force me to bet 20%.

Dunder,how much times you bet 20%,honestly?

I can't answer for Dunder but I have personally wagered 20% and more of my bankroll many times. Remember, when a Kelly wager is 20% or more of your bankroll, you are not wagering on 50-50 propositions, you are wagering on propositions with very high win rates. All of my wagers that had been at 20% or more were all some kind of hedge with very high win rates. I now only wager half-Kelly and very rarely at or over 20% of bankroll. I did not like the wild swings that full-Kelly produced and am satisfied with a slower growth rate and not deal with the more extreme ups and downs of full-Kelly. I also took a couple of nasty hits last year when Matchbook was having its problems and I couldn't close out a couple of hedges. This has also made me leary of having too large a % of my bankroll at risk at one time. Matchbook did however give me a $50 commission credit.

Now, Kelly does not "force" you to do anything. It is a tool that can be used to enhance your wagering once you understand the concept. I use Kelly to tell me what my max wager should be. Look at all the threads that throw around some "safe" fixed % to wager. With Kelly, once you determine your risk/win info, you will know what the max wager % should be. To say that 3% is a good % without knowing the details is wrong. Based on your particular details the max % may be 2%. Realize that even if you have a "winning" system you can still lose if you use a % that overbets.

Another thing that Kelly does for me is to help me determine which wagers to concentrate on. More often than not, you are better off playing the +EV wagers that have a higher winning % even though they may not have the greatest edge. Kelly helps me to determine which wagers have the higher expectation of growth.

It is my opinion that one will become a better gambler if they understand Kelly even if you don't use full-Kelly.

Joe.

**Shelton** · 02-14-10, 06:21 PM

D...wouldn't touch it

**Dunder** · 02-14-10, 07:27 PM

Originally posted by tachi

Kelly,Melly,Selly..

no system can force me to bet 20%.

Dunder,how much times you bet 20%,honestly?

Twice, both nerve wracking heavy chalk bets which came through.

If you read my response to the OP however, I actually suggested leaving this one alone despite what applying Kelly Criterion would say. Nobody is forced to do anything and I don´t actually use Kelly albeit the way I allocate units uses the same principles.