MLB Alternate Totals Lines

**KVB** · 08-01-17, 11:40 AM

Sure, there are always potentially good wagering opportunities available by moving off the Totals but over the long haul you will be paying too much in vig to likely make it profitable. Watching 2 games in one day won't get you very far.

If it's just about winning today's bet, if you do win, you will be shorted odds. It's better to try to get a good opener and let the rest of the market pay in number and price as the line moves your direction.

Over the long haul this will lead to profit, and getting limited.

Anything else is really considered a proposition and they are not usually well priced. You could lay a high price to move off the number a ways, but one loss sets you back a bit. Buying and selling off of the Totals isn't going to get you very far long term. If you knew which direction to buy, then just trading the offered line at reduced prices will get you much further.

Good Luck.

**Waterstpub87** · 08-01-17, 12:30 PM

Agreed. Consider that their are two ways to beat books: Beating them on Pricing, and Beating them on Value.

Beating them on Price: For example, betting correlated parlays, smart teasers, mis-pricings on alternative run lines, 1st halfs vs total games ect, chasing steam. This methods are easy to detect and to counteract. It is easy for them to look historically about how much 2 runs are worth in an MLB total, figure out the correct price, add a substantial margin, and then offer an alternative run line. .

Also, if you exploit a hole like this, it will get fixed quickly. Often times, if your only action is things like chasing steam and betting off numbers, especially on props, you will quickly become limited.

Beating them on Value: This is being able to set better odds than the bookmaker, and if you want to be a winning bettor, this is what you should strive for. You need to be able to set a more accurate line than the market open. This will be confirmed by the odds moving in your favorite after you bet.

**koreatine** · 08-16-17, 10:58 PM

Thanks for the PM Bsims

I cannot respond yet though. I don't have enough points. I will as soon as I do.

**HeeeHAWWWW** · 08-17-17, 05:46 AM

Originally posted by Bsims

1. Do the lines move in lock step? In other words, when the primary line moves, do the others automatically move by an equivalent amount? Or, do they move independently of the others?

Some books will move in lock step. Those with more sophisticated trading models will move more on the particular line with extra liability.

As for value, I often find them in the alt lines. Usually it's just a case of getting a little more EV - ie the primary line is value, but the alt lines might be better. Less common, the primary line is accurate but the alt lines aren't (because the relationship between different lines is sometimes steeper/shallower for different matchups).

**Snowball** · 08-18-17, 03:57 PM

The line adjustments are programmed, cookie-cutter and set..

but what makes buying points on totals a real strategy is to find
the games where the value exceeds the adjustment. Finding them
is what makes the adjustment inadequate to compensate for the
value in real terms. For example, two teams with strong offenses
and weak defenses will generate a higher total, and two teams
with strong defenses and weak offenses will generate a lower total,
but the point adjustments are the same - in reality, they are not the same.
in there is the edge. you still have to cap and watch out for injuries.
this tends to work best in basketball and football, however, not MLB.

**evo34** · 08-19-17, 01:51 AM

Originally posted by Bsims

In another thread, a new user asked a question about totals. This prompted the typical totals suck or totals great responses, but nothing enlightening. One guy did mention that he thought totals were a less efficient market possibly providing some opportunities. This prompted me to re-visit some thoughts I've had about alternate total lines.

Most books will give you a total value with a line attached to it. For instance TBA-HOU 8.5 +101 -124. But they also give you some alternative total lines like TBA-HOU 8 -113 -103. One book takes it to extremes, TBA-HOU 4.5 -841. +512. This leads me to two questions.

1. Do the lines move in lock step? In other words, when the primary line moves, do the others automatically move by an equivalent amount? Or, do they move independently of the others?

2. Are there potentially good wagering opportunities available by moving off the primary total? It's obvious by the lines, you pay a higher premium (vig) by moving away from the center. But if the market is indeed inefficient, are there some potentially hidden gems in here?

If you really know the answers, please share them. I've decided to monitor 2 games today at various times to see what I can glean from the movements. In particular, I've chosen TBA-HOU and SFG-OAK. These 2 had similar money lines and totals values at 7 AM this morning. I'll track their movements every couple of hours to see if the changes tell me anything.

Short answer: books are not necessarily perfect in pricing alt totals. The problem with looking for mispriced lines is that you have to be extremely sure of your own model. It's actually very tricky to get right in baseball, as bullpen usage patterns and player and team volatility all come into into play when predicting the odds of an extreme total. Even without using that level of complexity, you can find plenty of badly priced alt lines at 5 Dimes. Finding useful limits there is another story.

**Bsims** · 08-22-17, 03:52 PM

I started this thread when another poster postulated that he thought the alternate lines weren't overly efficient. The model I use is capable of generating a large amount of likely scores for any game. So I added code to look at the probability of hitting an alternate total and its expected return. It found a handful of games each day. S I decided to place some wagers while I had more time to analyze these lines. I'm almost certain that there isn't any source of historical data on these lines. Thus back testing isn't practical.

I asked the question to potentially get a better understanding of how these alternate lines behaved. Initially I looked at a couple of games, but that wasn't particularly enlightening. I thought it foolish to wager on untested system, but the results were good so I continued. I did add code that saved the alternate total lines several times a day.

I finally got a chance to look at these files and the results were discouraging in terms of being able to replicate these with historical data. One book appeared to be stable with 20 cent lines away from the center. But when I took a specific example, it was interesting. There did not appear to be a formula. For instance, when the middle total for the under 8 was -110, the under lines for 7 varied from -154 to -165. Not an obvious relationship.

I still have the data and may try more analysis, but for now I'll just trudge ahead. Just for the heck of it, here are today's picks (always a kiss of death sharing them).

Pick	Line	Prob
Under 8 oakbal0	334	31.2%
Under 8.5 cubcin0	246	34.2%
Over 11.5 colkcr0	262	33.1%
Under 7.5 nyydet0	205	37.1%
Under 7 boscle0	193	38.0%
Under 7.5 texlaa0	205	36.3%

**Bsims** · 08-23-17, 07:08 AM

Originally posted by Bsims

Pick	Line	Prob
Under 8 oakbal0	334	31.2%
Under 8.5 cubcin0	246	34.2%
Over 11.5 colkcr0	262	33.1%
Under 7.5 nyydet0	205	37.1%
Under 7 boscle0	193	38.0%
Under 7.5 texlaa0	205	36.3%

0-6. Yep, kiss of death by sharing, LOL. However, it did demonstrate something valuable. Sample size. My fear of course is that my success thus far is more the result of limited sample size rather than the results of something more solid. Thus the imperative need for back testing.

As I learned yesterday, generating historical alternate totals lines from primary historical data won't be simple. In the one book's case, there won't be a simple formula. Most likely what I would do is look for the combination of lines for each total that occurred most frequently. But there are 2 problems with this. First, it isn't easy and will take a fair amount of work. Second, this might introduce some bias. Maybe the book has good reason for changing "the formula" for some games and the back testing wouldn't account for that. So what to do? Here's what I will do.

1-Stop the major effort at this analysis for baseball now (since the season is almost over). Move on to looking at similar approaches in football and basketball. I'm not overly optimistic because these are more difficult sports to analyze.

2-Continue to wager on these through the end of the season. Although I could stop if I fall behind. This is a common money management approach guaranteed to make you a loser.

3-Continue to gather this alternate totals data for analysis this winter. I still have some macro questions that I might find time to look at. If I do, I'll share the results with you.

**Bsims** · 08-23-17, 06:40 PM

Today I had a chance to observe some totals line movements from the book I discussed yesterday. They only offer 5 alt totals lines, the primary and +/-0.5, and +/1 1.0 from the primary. I did observe a few things that I was interested in learning (again about this particular book only).

1-They always maintained a 20-cent line for all 5 totals.
2-They did not have a specific formula for computing away from the middle.
3-When the middle line moved, they all moved. But not in the same amount.

Here is the data for the Cub/Cin game tonight.

Time	Total	11	10.5	10	9.5	9
952	Over	143	115	-105	-125	-158
1406	Over	140	110	-110	-130	-160
1412	Over	145	115	100	-115	-150
1903	Over	150	122	-105	-125	-152

Time	Total	11	10.5	10	9.5	9
952	Under	-163	-135	-115	105	138
1406	Under	-160	-130	-110	110	140
1412	Under	-165	-135	-120	-105	130
1903	Under	-170	-142	-115	105	132

Look at the 9:52 and 19:03 lines. They are the same in the middle, but different away from the middle. This tells me 2 things. Again, coming up with a formula for generating the alt runs from a historic primary value is not possible.. But second, the differences might provide some wagering opportunities.

**Bsims** · 08-23-17, 07:17 PM

OK, now I’ve had a chance to look at another book for the same Cub-Cinn game. Here is the table and the differences from the first book.

Time	Total	13	12.5	12	11.5	11	10.5	10	9.5	9	8.5	8	7.5	7
724	Over	286	218	198	164	140	110	-103	-123	-161	-198	-251	-284	-458
952	Over	286	218	198	164	140	110	-103	-123	-161	-198	-251	-284	-458
1406	Over	271	208	189	157	133	105	-108	-129	-170	-208	-265	-300	-489
1412	Over	286	218	198	164	140	110	-103	-123	-161	-198	-251	-284	-458
1903	Over	271	208	189	157	133	105	-108	-129	-170	-208	-265	-300	-489

Time	Total	13	12.5	12	11.5	11	10.5	10	9.5	9	8.5	8	7.5	7
724	Undr	-392	-285	-256	-207	-174	-135	-113	100	130	157	195	217	325
952	Undr	-392	-285	-256	-207	-174	-135	-113	100	130	157	195	217	325
1406	Undr	-369	-270	-242	-197	-165	-129	-108	105	137	165	205	228	343
1412	Undr	-392	-285	-256	-207	-174	-135	-113	100	130	157	195	217	325
1903	Undr	-369	-270	-242	-197	-165	-129	-108	105	137	165	205	228	343

1-More totals (there are more, just couldn’t fit them on the page).
2-Terrible vig as you move away from the center.
3-Movement appears in lockstep.
4-It does appear as though one could develop a formula for the off center lines. A different one would have to be developed for different primary totals.

Numbers 3 & 4 are different than what I thought I saw earlier. So more study is needed. But that’s enough for now.