Here’s a fun one…big and broad sample size with fairly straightforward plain English, but perhaps a complicated looking SDQL code. Let me break it down:
H and total-p:total<=0 and total-(tA(total)+oA(total))/2>0
In database history, this is O/U:1890-2232-44 (54.1% for the Under).
Note: I don’t have exact figures, but I believe Unders tend to carry the chalk more often than overs in NHL so you may just want to keep this one in your back pocket as a filter or a starting point for a wager on an under. On the other hand, when you hear the Plain English explanation, I think you might agree that this is sort of a contrarian angle where you might actually be getting plus odds or small chalk. I would have to refer to another database to get the actual roi and won/lost $/units on this one.
Let’s say the odds were -115 though for every one of these total wagers. $100 bettors would be up $5,850 (pretty low roi, but we’re only making either a conservative assumption or maybe a little too generous one as to what the actual odds were on these wagers; one day we’ll get lines in the database). Enough waffling though….
HERE’S THE PLAIN ENGLISH DESCRIPTION:
1. ‘H and total-p:total<=0’ — This part is saying that today’s line for the home team (home team to avoid duplicate results….although, do note that results are slightly different if you choose the road team instead or the Dog or the Favorite which would also eliminate duplicates) is LESS THAN that team’s total from their previous game.
2. ‘… and total-(tA(total)+oA(total))/2>0’ — This part is saying that if you add and divide by two the average season-to-date total per game for each team in the matchup at hand, and then subtract that number (which is sort of a basic model you might use to predict what the line will be if given only the names of two teams) FROM today’s total, and that figure is positive — meaning, today’s total is LARGER than the expected total based on the big picture….whereas, parameter one is meant to lure in the square gamblers who never look further beyond what happened yesterday: aka. “24-hour-itis”
So in a nutshell, I feel like you might be getting an average line over something like -105 or maybe even better here since a trend in all sports amongst the masses tends to be decided where value is based on information from yesterday….and that folks is the ground work for a good ole “trap”
Here is a slightly different version of the system with better results, but perhaps less simplicity (and therefore, possibly less reliable). Who knows though? Time will only tell; always fun to come back to these systems years later and see which ones flourished and which ones failed. That is how you learn…
24-Hour-Itis NHL UNDER SYSTEM VER. 2.0
H and (total-p:total)<(total-((tA(total)+oA(total))/2))>0.1
The Under is 1147-893-38 56.2% now when the current total minus the home team’s previous total is less than the total minus the season to date average vegas totals for both teams combined. In other words, the total is smaller today than yesterday’s but larger than average for the two teams’ average line. Same exact idea: people will be be skittish on a seemingly worse deal than yesterday instead of looking at the big picture which is that the current line should actually be higher.
Note: the “>0.1” part is there to say that today’s total is, somewhat arbitrarily about 2% larger than expectations based on season to date results.
Why 0.1 though? No particular reason for that specific number; however try running this grouping:
H and (total-p:total)<(total-((tA(total)+oA(total))/2))>0,.01,.02,.03,.04,.05,.06,.07,.08,.09,.1,.11,.12,.13,.14,.15,.16,.17,.18,.19,.2
Now sort by the blue ‘SDQL’ header on the right: perfectly linear results as the current total gets larger and larger than the expected total which directly strengthens the whole premise.