Thread: Optimal Betting
View Single Post
Old 01-08-2020, 07:51 PM   #10
rdiam
Grade 1
 
Join Date: Sep 2016
Posts: 606
Raptor:

You are confusing variance with edge. Since the edge in most gambling games is quite small (usually 5% or below), the variance tends to be high. High variance means that you can show a profit over a relatively small sample size but still be betting into -EV and having a negative edge. In fact, the whole idea of casino games such as craps, roulette, and slots, is the provide a positive edge (+EV) for the house and negative edge (-EV) to the player, but all showing a high variance so that in the short run there will be winners! If you win at craps or slots, you will be inundated with offers for meals and rooms to get you back in the casino. Why: because they know over a large enough sample size the player's negative edge will work in their favor and they will get the money given enough time.

The best gambling games are those with high variance but where those with skill can win over a large enough sample size. This is why lowball poker died (variance was too low and skilled players always got the money) and why no-limit hold em has become so popular (high variance, poor players can win in the short term, but skilled players can have +EV with proper playing and betting strategy). It is also why casinos try to ban blackjack card counters and shuffle trackers, since they know these types are playing with a positive edge, and will eventually win if they are allowed to play long enough. Similarly, most of the sportsbooks now legal in the US will limit or ban those bettors who bet with a pattern that shows them to have +EV. So if you are allowed to bet whatever your want with a sportsbook, you most likely have -EV since they welcome your action.

Since this is a horse racing site, consider that horse race betting also has high variance, especially in the exotics. Yes, you can have +EV and an edge, but having an edge has nothing to do with short-term results. And because of the high variance, with sample size needed for you "edge" to kick in is much higher than you might expect -- we are talking hundreds, if not thousands of bets.

Richard
rdiam is offline   Reply With Quote