[doc]

All right SBR I have a question for the bases cappers on here.

How many of you keep track of your action in excel?

For those who apply, it might be prudent to go through this year so far and analyze the moneyline faves you took.

Compare the games you WON to the score of the game... Did it cover the 1.5 RL?

Obviously all of your ML favorites will not cover the number, but the analysis I am doing for a few friends (with shocking results) is whether or not the money they lost by laying -150 as opposed to laying 1 unit to win +125 is worth the risk?

In other words when you play straight run lines with positive vig you always risk less to win more as opposed to ML faves where you risk more to win less.

Anyone follow me here? Any thoughts?

[EBone]

I understand what you're saying and would be interesting to know. However, for me, since I don't play many favorites, this would not work. However, the concept would work if I used the ALT -1.5 Runline. That actually would be interesting but I don't know where to find historic alternate runline info.

Since I'm negative for the year, I guess I better start figuring out how to maximize the positives.


E

[Ganchrow]

All right SBR I have a question for the bases cappers on here.

How many of you keep track of your action in excel?

For those who apply, it might be prudent to go through this year so far and analyze the moneyline faves you took.

Compare the games you WON to the score of the game... Did it cover the 1.5 RL?

Obviously all of your ML favorites will not cover the number, but the analysis I am doing for a few friends (with shocking results) is whether or not the money they lost by laying -150 as opposed to laying 1 unit to win +125 is worth the risk?

In other words when you play straight run lines with positive vig you always risk less to win more as opposed to ML faves where you risk more to win less.

Anyone follow me here? Any thoughts?

This is certainly an interesting topic, but I'd just caution you to make sure that your sample set is large enough to draw meaningful conclusions. This might be a bit problematic if you're looking at only the games on which your friends have wagered this past season. What I always suggest is to come up with a theory first, test and minimally calibrate it on some old data, and only after that proves successful, gauge how you might have fared for the current season.

On the modelling side of things, a prerequisite for determining the fair valuation of a run line relative to a money line is determining the probability of the favorite winning by exactly one run (once you've done that coming up with the RL is just algebra). At a bare minimum you'd have to consider this for various totals as well as for home versus away favorites. (All else being equal, the probability of an exactly one run fave win is higher both for lower totals and for home faves.)

After that, you'd probably also want to look at magnitude of favorite-dom. All else equal again, we really shouldn't expect the same percentage of a -120 favorite's wins to end in one run victory as a -280 favorite's. Most books, btw, don't really handle this well -- if at all.

If you wanted to go a step further, you also might want to consider this for different flavors of games. Just to pick a couple of examples (all else equal of course):

1. We would expect a game featuring a dog with a stronger bullpen to have a higher probability of a one run favorite victory than a game featuring a dog with a stronger starting pitcher.
2. All else equal, we would expect a fave comprised of big power hitters, to produce fewer one run victories than a fave comprised of speedier ground-ball hitters.

What these two examples of course have in common is volatility. The more volatile the expected outcome of a game the more valuable a given fave RL becomes. This is an important and often over looked consideration, both when evaluating run lines versus money lines, and when evaluating the purchase or sale of points on totals.

[RickySteve]

(All else being equal, the probability of an exactly one run fave win is higher both for lower totals and for home faves.)

Given same run environment, a small home dog can be more likely to win by 1 run than a very large home favorite.

[Ganchrow]

Given same run environment, a small home dog can be more likely to win by 1 run than a very large home favorite.

The way I'd put it could be that we might expect a greater percentage of the small home dog's victories to be decided by exactly one run than the large home fave's victories.

Check out the Pinnacle Pulse where Simon discusses this very issue.

[RickySteve]

Given an expected total score of 9, a -264 home favorite would win by 1 run less often than a +111 home dog.

[Ganchrow]

Given an expected total score of 9, a -264 home favorite would win by 1 run less often than a +111 home dog.

While it might seem counter-intuitive, if you go through the historical record you do consistently find that all else being equal, as a team's probability of victory increases its probability of a one-run victory increases as well.

Just looking at four games I pulled at random off BigGuy.com:

Big Faves

7/2/2006 - 9½ Total
PHI	+200	+1½ -105
TOR	-220	-1½ -115

16.26% chance of a Fave victory by exactly 1 run.

6/18/2006 - 9½ Total
BAL	+210	+1½ +100
NYM	-230	-1½ -120

16.19% chance of a Fave victory by exactly 1 run.



Small Dogs

7/2/2006 - 9 Total
BOS	-130	+1½ -250
FLA	+120	-1½ +210

13.46% chance of a Dog victory by exactly 1 run.

7/2/2006 - 8.5 Total
SEA	+110	-1½ +190
LOS	-120	+1½ -220

13.21% chance of a Dog victory by exactly 1 run.

[RickySteve]

The likelihood of a home team winning by exactly 1 doesn't increase unbounded as probability of victory increases. It reaches a peak at around -150 and declines thereafter. So, there is a point of intersection in which a home dog has a greater probability of a 1-run win than a large home favorite

Home %1-Run
Odds Win
----- -------
130 16.07%
120 16.32%
110 16.57%
100 16.79%
(110) 16.97%
(120) 17.08%
(130) 17.15%
(140) 17.19%
(150) 17.19%
(160) 17.18%
(170) 17.14%
(180) 17.09%
(190) 17.02%
(200) 16.95%
(210) 16.87%
(220) 16.78%
(230) 16.69%
(240) 16.59%
(250) 16.49%
(260) 16.39%
(270) 16.28%
(280) 16.18%
(290) 16.07%
(300) 15.96%

[RickySteve]

The above table is for an expected total of 9.