Hi all,

I'm starting to look in to middles and sides as a way of betting on sports, but am trying to understand how to calculate the odds

If there's a middle on a game of NBA, -9.5 and +10.5, both with odds of $1.90(-111). If I put $100 on each, I'm losing $10 if the middle doesn't come off, but making $180 profit if it does. So if risking $10 to win $180, you could say the odds I'm getting on the middle are 18-1.

If there's a side on a game of NBA, -9 and +9.5, both with odds of $1.90(-111). If I put $100 on each, I'm losing $10 if the side doesn't come off, but making $90 profit if it does. So if risking $10 to win $90, you could say the odds I'm getting on the side are 9-1.

What if there's a middle and a side as part of the one bet, say -9 and +10.5, both with odds of $1.90 what do you think the odds on that total middle/side combination are? I'm still risking $10 in total if it doesn't come off, but I'm a bit confused as to what my potential payout is in this case. It could either be $90 or $180.

I had 4 thoughts here:

1. Do the odds just get added together? So the combined odds of a 18-1 chance and a 9-1 chance is 6-1.

2. Should the bets be treated separately and instead of $10 risk on each payout, because there are 2 options, it's really $5 risk on each option, which would result in 36-1 odds for the middle, and 18-1 for the side. Combining these together is 12-1.

3. Should I be average the potential payout? So $180 + $90 is $270. Halve that, which is $135. So I'm risking $10 to win an average of $135, so 13.5-1?

4. Following on from option 3, is it too simple to just say average the potential payouts. It might work out ok on an NBA game because the chance of either margin is probably very similar, but what if this was NFL, where the chance of 10 would be greater than the chance of 9. Does the chance of each payout occurring need to be factored in?


Any guidance would be appreciated.

Thanks

DK

your math is fundamentally correct but you're not applying it properly

ie unless you can accurately know how frequently you will succeed then any of your roi thoughts are pretty meaningless

you're effectively treating it as if it'll succeed half the time when common sense indicates that's not even remotely possible, you're also confusing payout with probability, just because you need to hit 1/9 to break even doesn't mean that's what the odds of it happening are - vig is a real thing

a good exercise would be to go through a database and find all of the nba middles you would have done the last 5 years running and then see what your profit/loss would have been

from there you could see it works and go on and print money or perhaps narrow the scope until you find a certain threshold where it is profitable

most importantly, if you were to search this forum for other posts of people trying to win by middling you'll end up finding a lot of posts by novices like you who think they found an exploit and very little in the way of the dozen or so professional sports bettors who do or did populate this forum

but having said that, line shopping is very much a key to success, i'm sure good middling opportunities present themselves but it's more likely due to getting very talented at scripting to scout out bad lines and execute a bet on them and/or just having an early bet on your preferred side and then being able to hedge out a middle later on in the week once vegas adjusts to the sharper side of things you were on from
the get go

gl and please share your journey here if you continue down the quest for middles sb could use some new life these days

That sounds like you think he thinks winning the bet is 50/50. What he’s assuming is that the number landing on 9 is just as likely as the number landing on 10. He may be wrong.

No.

Kind of. I prefer looking at is as probabilities applied to each payout. In your example that would be .5*$180 + .5*$90 = $90+$45 = $135. A better approach would be if you know or can reasonably estimate the odds of each number. Let’s assume there is a 1/20 chance of the number landing on 9 and a 1/21 chance of the number landing on 10. So $90/20 + $180/21 = 4.5+8.57 = $13.07 (which is more than $10, so a profitable bet in the long run if the estimates are correct).

Yes. I see on one chart where NFL games land on 10 about 5.5% of the time, and on 9 about 1.5% of the time. So using the numbers in your example, .055*$180 + .015*$90 = 9.9+1.35 = $11.25. But there are two other factors to consider. The first is that those percentages were on games ending on those numbers, not the favorite winning by those numbers. Obviously, the favorite will win by large margins more often than the underdog, but not enough to make that wager profitable, in my opinion. The second factor is that the number landing on either 9 or 10 is probably far more likely for a 9 or 10 point favorite than the average NFL game. So that works in your favor. Is it enough to overcome the error in the first factor I mentioned? I haven't a clue. What you really want to find in the NFL are middles including a 7, or better yet, a 3. Those percentages are about 9.5% and 14% respectively, and probably profitable even without the extra half point.