Using 2H bets to middle in basketball

[MrLomez]

New guy here. I'm interested in people's opinions/advice on how best to use 2H bets to create middles from existing bets on the opposite side.

(That sounds confusing...) I'll use an example to illustrate what I mean:

On Saturday I bet STL/MICHSt u124. The halftime score total was 47. At halftime I bet 2H o66. This left me with a significant middle (any number between 113 - 124 would win money, with both bets hitting on 9 of the numbers), and a guarantee to win at least one of the bets.

A second example from the weekend was a bet I made on Lehigh +5 over Xavier. Lehigh was up 4 at half. I then bet Xavier -4 (2H), opening a 6 point middle between Xavier losing by 1 and winning by 5 or less.

Even though neither of these middles came in, they intuitively seemed like smart plays (to me, anyway), though I'm having a hard time corroborating my intuition with any kind of mathematical proof.

Would you guys make these plays? Why/why not? And does anyone know the math you would use to determine when to go forward with a play like this, versus when to just stick with the initial bet? (Assuming all bets at -110)

Thanks. Looking forward to people's responses.

[lindetrain]

How to best use them: don't use them.

As BTB will probably come in here and tell you, only time you should really middle is when you have a ML wager that is giving you pretty decent odds, and you can middle that ML with a 2H point spread and still profit if the middle doesn't come in (i.e., if you have a +300 ML dog who is up by 12 at half and 2H wagering opens up a 4-point middle between your ML bet and the 2H bet). Middling a -110 wager with a 2H -110 wager is not the way to go; don't give the books an out and a chance to get your money back.

That being said, there is one small exception. If you are ABSOLUTELY POSITIVE that the 2H wager is a strong wager in its very own right, it may be worth considering and playing for half your initial wager. I've had a couple occasions where I've been holding what appears to eventually be a winning ticket at the half, but have seen something in the game that's convinced me the other side makes for a great 2H wager. I'll middle for half, only because I really really like the 2H wager as its own play.

[atk825]

I am of the thought that you think of a second half bet (the middle bet) as an independent wager. If you feel that there is value in that wager, then go ahead and make it. If not, then don't.

In this case, do you think that Xavier had a good chance at winning the game outright? If so, then make the bet, it should be irrelevant whether you have Lehigh +5, Xavier -5 or no bet on the game. Being able to win both bets is just an added bonus.

[Beyond the Bets]

I agree with both lindetrain and atk on this one. Middling is often times a waste if you're merely trying to slip through a tight window. For instance, let's say you took the Knicks -9 yesterday, and they led by 9 at the half.

Second half line: Knicks -2.5.

Do you want to shoot for a Knicks 10 or 11-point win to hit your middle, or do you merely want to keep your original bet? If you still have confidence in the bet after watching the first half, then you'd probably want to maintain the advantage you've earned. The Knicks are a 2.5-point favorite in the second half, which implies that they're expected to win by 11.5 overall.

You're theoretically in a good position to win your original bet.

However, let's say you've watched the game, and you're leaking confidence. You no longer like the Knicks and don't think they'll hold on. Well, if that's the case, then maybe you want to attempt a middle. It's a way to hedge your original bet and to protect yourself, and at the same time, perhaps you'll get lucky and hit both.

In any case, I'm rarely a fan of middling if it ends up taking a win off the board. You start turning sure wins into splits (and no profit), and you're asking for trouble, particularly when you go on a bad run.

[MrLomez]

Thanks for the quick responses. I've spent most of today looking around for a way to bear this out mathematically and have finally found a way to crunch the numbers (perhaps crudely). As it turns out, unsurprisingly, you guys are right.

I may be talking to myself here, but the short of it is that a 2H bet to attempt a middle will yield, on average, between 5% - 8% LESS than no bet at all. This is true even if you can get a 2H bet at discounted or no vig. (Interestingly, the yield difference is biggest at both extremes, when there's either a wide middle or if you're dealing with 1/2 point. It settles to about 5% when the middle is +/-5 the expected number.)

If anyone cares about the math side of it, I'd be happy to share.

/lesson learned

[Beyond the Bets]

What's your background, Mr. Lomez?

Math geek? Accountant?

Statistician?

[VincentValour]

(03-21-2012 03:48 AM)lindetrain Wrote:  How to best use them: don't use them.

As BTB will probably come in here and tell you, only time you should really middle is when you have a ML wager that is giving you pretty decent odds, and you can middle that ML with a 2H point spread and still profit if the middle doesn't come in (i.e., if you have a +300 ML dog who is up by 12 at half and 2H wagering opens up a 4-point middle between your ML bet and the 2H bet). Middling a -110 wager with a 2H -110 wager is not the way to go; don't give the books an out and a chance to get your money back.

That being said, there is one small exception. If you are ABSOLUTELY POSITIVE that the 2H wager is a strong wager in its very own right, it may be worth considering and playing for half your initial wager. I've had a couple occasions where I've been holding what appears to eventually be a winning ticket at the half, but have seen something in the game that's convinced me the other side makes for a great 2H wager. I'll middle for half, only because I really really like the 2H wager as its own play.

Hit it perfectly. I love second half wagers in their own right. Many games I'll look at and think that is a 2nd half wager waiting to happen. There are a few specific instances where I really enjoy them. Usually using the adjusted game total.

# 1 Big home favorite is behind at half.
In most sports betting big favorites is not a long term winning strategy, the lines are usually shaded by a point. So if a big favorite is down 4 at half being a 10 point game favorite, and the 2nd half total is only -8 bringing the total game total to say -4 there are 6 points of value gained by taking that favorite.

# 2 Same concept as # 1 using totals. Two low scoring teams have a shoot out in first half, value should be on the under in 2nd half. Same with high scoring teams scoring a low amount.

# 3 Using key numbers to your advantage in 2h betting. Like taking a trailing home favorite where the adjusted game total ends up -2 or -1. Odds are because of foul shots, if the home teams comes back and wins straight up they will cover the two points. So this gives the value of the adjusted game line and value on two points that are mostly dead numbers. Works also in football with 3, 4, and 7.

[The Cosmic Kid]

I successfully middled the Gonzaga-Ohio State game. It was a fluke with a pair of free throws with .01 seconds left.

I actually bet the second half more than my original bet because I doubted my original bet and felt like Ohio State had a run in them. I was wrong, but got lucky and won both.

[MrLomez]

(03-21-2012 06:03 PM)Beyond the Bets Wrote:  What's your background, Mr. Lomez?

Math geek? Accountant?

Statistician?

Though I'm currently far afield from anything numbers related, I worked in finance for a spell, then for a certain company in Mountain View, CA (unfortunately I don't program). To be honest, most of the stats stuff relevant to sports gambling, so far as I can tell, is at best a dim memory from undergrad. In short: I KNOW NOTHING.

And since we're on the subject of basketball totals, does anyone know how game totals and 2H totals are adjusted to account for the possibility of overtime? Or is this a book by book, capper by capper sort of thing?

[Beyond the Bets]

As a general rule, I'll never bet a second half under in a low-scoring game that's extremely close and figures to stay that way. For one, you've got the potential for late-game fouls which can send the game over by itself. Second, the game could go to overtime, which would be a dagger.

I'd much rather play a second half under on a blowout game where one team lets off the gas and uses up clock.

[r2edline]

Although I don't really condone middling, I have had some recent success playing 2H lines in spots where I'm getting extreme value on the point spread.

If I can get 10 or more points of value on the 2H line compared to the spread pre-game, I'm going to take it most of the time. i.e. Team A is a 6-point fav over Team B and has a 18 point lead at halftime. The halftime spread is Team B -2 (adjusted game spread of Team B +16).

My theory is that even though the first half was a blowout, the original point spread created by the bookmakers is going to be fairly accurate +/- 10 points most of the time. It's been really successful for me throughout conference and march madness tourneys.

Sometimes it works out as a middle if you have a pre-game play (more often on totals), but if you have Kentucky -5 and they're up 20 at the half, there isn't much of a point middling as you're probably going to win anyways.

If I was looking to employ this strategy on a middle attempt for a total, I would want 15-20 points of line value to give you a decent chance - no point in attempting a middle for 5 or less points. You're basically just hedging.

[MrLomez]

r2edline,

I think what you're describing is the same logic trap that I was falling into when I first posted this--i.e. the original line, regardless of what happens in the first half, should maintain a ~ +/-10 variance for a spread (maybe closer to 20 for a total). Ergo, if a 2H line falls outside of that variance, it's a strong play to take the side that leans toward the original number.

So far as I can tell, this, in itself, will not produce a +EV.

Here's the basic problem with this kind of bet, as I see it:

As you say, the underlying basis for such a 2H bet is your faith in the bookmaker to set a strong opening line. That same bookmaker, based on the same model (presumably) he used to craft the opening line, also sets the 2H line. So if you believe in the game line, why, for the 2H line, would you assume that he's suddenly got it wrong? Moreover, if the bookmaker is sharp, wouldn't the 2H line reflect the tendency for numbers to move back to the center (assuming such a tendency even exists), or reflect the fact that an overperforming dog will come back to earth, or a favorite with a big lead will play subs, or whatever?

To put it neatly: if a bookmaker makes a sharp total line, he'll also make a sharp 2H line, and your likelihood of beating either is the same.

This analysis, of course, may well be way off. Perhaps there is in fact a bias in 2H numbers to overreact to 1H's. Is there, that you know of?