Question about betting favorites

[rjp]

[ QUOTE ]
Well, right, if you bet less on a +EV wager you're leaving money on the table... but that would be equally true of any bet you make. I guess you left me several levels behind with that point.

[/ QUOTE ]

The point is that either way you look at it you've got an expected return of 5% (for the sake of the argument). If you wager more at -250 than you do at -110 then (in relation to your bankroll) your actual return is higher for the -250 than it is for the -110. If you have a small bankroll that isn't suited well for the random walk of a ~55% expectation then you might be correct in wagering more at -250 over -110 because you have no choice, and you might as well not bet the -110 at all if you can't sustain the random walk. If, however, you have a bankroll that is able to sustain the random walks then you're cutting yourself short on the -110.

[Thremp]

rjp,

Okay, now I'm sure. That isn't correct. Plug it into Kelly Criterion. You are not "shorting yourself" out of any EV or expectation, you'd have to increase RoR to gain more EV with the gain in wager size etc etc.


Colonel,

Essentially yes. Usually you'll be betting a whole slew of these so you can change your bet sizes and also some hedging aspects come in (Futures for example).

[fnurt]

[ QUOTE ]
If you have a small bankroll that isn't suited well for the random walk of a ~55% expectation then you might be correct in wagering more at -250 over -110 because you have no choice, and you might as well not bet the -110 at all if you can't sustain the random walk.

[/ QUOTE ]

I'm getting more and more confused by this point. The only reason we're talking about bet size in the first place is to avoid risk of ruin. We're starting with the assumption that if there were no such thing as odds, every bet would be 1-2% of bankroll, and then we're trying to adjust that concept for a world that actually includes odds, sometimes steep odds.

[rjp]

Bet size is about more than risk of ruin, it's about maximizing every dollar you wager.

My bankroll is used to bet anything from large money line underdogs to large money line favorites (and everything in between).

Is it better to wager 250 to win 100? It depends on what your bankroll is for. If you only have to withstand ruin on 75% bets with a 5% expectation then risk an appropriate amount of your bankroll, and I'd suggest plugging it into a kelly calculator as Thremp suggested.

For me, there are way more 5% expectations at -110 than there are at odds of -250 or higher, so my bankroll must be able a) withstand fluctuations of a 55% chance event and b) maximize every dollar I wager. As such, by risking X to win 1 at -250 I will limit my bankroll growth because I'm not making as much in relation to my bankroll on wagers of -110. If you find the holy grail and are able to make most of your bets at -250 odds or greater in which you've got a 5% expectation then let me know. [img]/images/graemlins/smile.gif[/img]

[hogua]

[ QUOTE ]
$250 to win $50

Kelly criterion suggests the same -- a -110 bet with a 5% roi, you should bet $550 to win $500 (with a $10k roll) and a -250 bet with a 5% roi, you should bet $1250 to win $500. So basically, your "to win" amount stays constant.

[/ QUOTE ]

This is correct, as long as the expected ROI is the same for each wager and/or you can't determine the specific ROI of a wager.

So, OP, to answer your original question, you shouldn't just look at the line. You need to take your ROI (your edge) into account as well.

Keep in mind that it can be difficult for even seasoned handicappers to determine a ROI for a specific wager. When you can't come up with a specific ROI for the bet, you can assume it will be the same as (or close enough to )your long term historical ROI for similar wagers.

Someone mentioned a case where you'd get -110 on something where the true odds should be -250. If you are confident that your getting that high of an edge, you should bet more than you would if your edge was only 5%.

[Thremp]

Note: What rjp was saying is right in the Kelly criterion your Ev is much higher when you wager more with a same edge.

[mynamewastaken]

Wowowow...

I thought I understood the original question, but after reading through this thread twice I'm more confused than ever.

Was this the original q:

unit size: $50... ie at -110 bet $55 to win $50, at +110 bet $50 to win $55.

Thus, in theory, if you're betting a -200, risk $100 to win $50.

At -500, risk $250 to win $50...

Do you ever lay $250? The answer to me has to do with if you can afford it. You can't bet above your bankroll, even though a fav bet with a positive expectation yields less variance, you can't afford to go beyond your roll.

Example: What if there is a ML of -20000 but you've capped it at -25000... Therefore there's value in making that bet, significant value at that, but losing once might wipe out your roll.

You're not leaving money on the table by passing up a +EV bet for the sake of risk aversion.

I pass on +EV opportunities all the time in the name of reducing RoR. Gotta pick your spots.

Can someone run down the other arguments presented here in a 'for dummies' version?

[fnurt]

Yes, the original question can be boiled down to "If your unit size is $50, should you always be betting to win $50?" People have been making some good points on related issues though.

[skierdude1000]

[ QUOTE ]
Yes, the original question can be boiled down to "If your unit size is $50, should you always be betting to win $50?" People have been making some good points on related issues though.

[/ QUOTE ]

and i don't think there is an answer to this question... seems like its up for contention based on the posts I've seen

[Jaguar101]

[ QUOTE ]
Yes, the original question can be boiled down to "If your unit size is $50, should you always be betting to win $50?" People have been making some good points on related issues though.

[/ QUOTE ]

How much is your roll, then the answer can be solved