Basic question on a spread

**Flight** · 11-11-10, 06:24 PM

Depends on your objectives and bank roll. Some would scalp +129/-120, others would play the soft side (-120).

Perhaps I'm missing what you're really getting at here?

**lordswing** · 11-11-10, 06:30 PM

Sorry I wasn't clear enough, my goal is to maximize my growth, therefore no hedging.

I want to figure out the % I should bet (ideally listening to Mr. Kelly, but I'm listening to other poster's thought process).

**u21c3f6** · 11-11-10, 07:19 PM

Originally posted by lordswing

Sorry I wasn't clear enough, my goal is to maximize my growth, therefore no hedging.

...

I am curious why you make the statement above. Actually, you can maximize growth through the use of hedging (assuming no undue constraints).

Joe.

**lordswing** · 11-11-10, 09:12 PM

Originally posted by u21c3f6

I am curious why you make the statement above. Actually, you can maximize growth through the use of hedging (assuming no undue constraints).

Joe.

I would only hedge when either both sides are better than market price, or I am uncomfortable with the size of my wager/future

Just because you're guaranteed a profit, does not mean you're maximizing it

**Blax0r** · 11-12-10, 03:44 AM

I'll give this a stab.

The first option is to hedge for a riskless 1.8% profit of your total scalp (~44% on the dog and ~56% on the fav), which seems like a decent percentage.

The second option is to bet on the soft side (-120); I am going to assume that the original lines are the "correct/true" odds and properly priced.

The Kelly-bet size should be the correct wager amount given that assumption (for once we know the probability of winning!), but I wouldn't hold it against you to apply a multiplier. Anyway, you should just compute the EV of of the -120 wager and divide it by the odds (5/6). That quotient is the percentage of your bankroll to wager (basic definition of Kelly).

Also, given that assumption, you don't want to put any wagers on the +129 line since it's theoretically priced correctly and should not be positive EV.

Anyway, the EV from the 2nd option should be much higher than the 1st option, BUT given the Kelly betting size constraint, you'll actually make more from just scalping (assuming you can scalp your whole bankroll and the probability of the scalp succeeding is 1). You can bet more than Kelly if desired in order to shoot for a higher payout, but it isn't optimal (I forget the specifics of the proof, but it's all there).

The betting behavior is the mostly the same for both scenarios you listed; I haven't thought through the effects of having in-game betting options available though. If possible, I would try to do both (you'll probably need a friend for this); apply the kelly-sized bet on -120, and scalp the rest of your bankroll.

**u21c3f6** · 11-12-10, 05:22 AM

In addition to Blax0r's post above, read this entire thread:

SBR Forum - Sports Betting Talk - Sportsbook Review Forum

http://forum.sbrforum.com/handicapper-think-tank/697291-did-i-mess-up-mnf-my-local.html

Sports betting and handicapping forum: discuss picks, odds, and predictions for upcoming games and results on latest bets.

Joe.

**splash** · 11-12-10, 10:00 AM

Lets say you have a bankroll of $1000.

If you arb, you can guarantee yourself about $18.18 profit by betting $555.38 on +0.5 -120 and $444.62 on -0.5 +129.

If you bet kelly, you need to calculate the no vig market line. The no vig line in this case is approx +/-134 with a win% of 57.3%/42.7%. See http://forum.sbrforum.com/handicappe...rcentages.html. Using -134 as the true line, you have an edge since you can bet at -120. Applying Kelly, we have ((.833*.573)-.427)/.833=.06132. So Kelly tells us to bet 6.132% of our bankroll or $61.32. If we do so, we win $51.10 57.3% of the time and lose $61.32 42.7% of the time. Which equates to an EV of (51.1*.573)+(-61.32*.427)= $3.10.

You're much better off arbing.

**lordswing** · 11-12-10, 12:54 PM

Hedging/arbing is not possible for me (can't put money into a Pinny acount), so let's leave it at that. I can only play the +0.5 -120, I would like to know what % is best to maximize odds.

I was originally just figuring out the juice and getting the non-vig % from that. In this case, -142/129 gets you 57% for the fav, and inputting -120 as the price I can get equals to a generic half kelly of 3% for me. This is very wrong I believe, as I'm not factoring in the spread at all.

Another thing I'm doing wrong, with Quarter/Period lines, the case of a tie is present. Thus, you need to multiply the win % (whatever the ML is, I've already forgotten but one can easily be made up) by the number of times a Quarter/Period ends in a tie to get your "true" Quarter/Period win %. Then using this as your win % with Kelly.

**splash** · 11-12-10, 01:00 PM

Originally posted by lordswing

This is very wrong I believe, as I'm not factoring in the spread at all.

Kelly has nothing to do with the spread.

**splash** · 11-12-10, 01:05 PM

The spread could be -13.5 +129/ +13.5 -142, as long as you're still getting +13.5 -120 the kelly calculation would be the exact same.

**Blax0r** · 11-12-10, 02:13 PM

As splash has already stated, the spread and Kelly aren't really related. The spread should only affect the win % if you are trying to divine your own probability figure.

I didn't think ties were possible given the .5 spread (in any quarter or game), but I agree that it's definitely something you should factor in if the spread is a whole number.

**lordswing** · 11-12-10, 03:27 PM

Would this be "correct" then?

I see Pinny is offering Stars 1P +0.5 -180/-0.5 +165. I believe this is an efficient market, stripping away the juice I assume this is the win % (-180/165 -> 63% non-vig odds).

My "local" is offering Stars 1P +0.5 -155 for some strange reason. Using the kelly calculator, half kelly, -155 for my odds and 63% as my win probability, I see 2.8x.

**splash** · 11-12-10, 03:33 PM

Looks good to me.