Better late than never – I suppose. The fact is, I’ve been running the model for a couple weeks – with rather poor results. Not sure where the problem lies because making fairly accurate score NBA projections, given more than 2 months of data, should be pretty much a snap; but that hasn’t been the case thus far. I’ll give it a few weeks, but there are a handful of red flag targets, which if they hit, I’ll just stop posting.
Not a red flag, as such, is a poor ATS record; more important is the projected margin of victory (MOV) which decides everything. Posting, as I will, a few hours before tip-off means that late line movements and game-time starter-sitting, make a joke out of early projected ATS winners - but those are outliers, and nothing to do with the model’s reliability. Although I track the ATS record, far more important is variation in the MOV which, when it consistently goes wide, is a major red flag.
Totals projections have never been that reliable, so caveat bettor.
The model is based on past performance team and player data, rotations, injuries, and game matchups. The first posting of the week will include a standings frame for reference, and the thread will be updated with daily projection posts. Bear in mind, these are math-based projections, not “predictions”, and should be used only as a guide to betting; and, again, the posted MOV is far more useful to a betting decision than the ATS “pick”.
The final ATS pick will be based on the posted MOV and the consensus closing lines at VSiN which, for record keeping purposes may involve pick-flips, that late spread line adjustments may require. The posted MOV doesn’t change, but the final pick always depends on the closing line.
As I said, I’ll be posting hours before tip-off for the earliest games, except for early morning tip-offs on some weekends. Residing at 19.89° N, 155.66° W, I won’t be getting up in the dark to do this. The projection posts will come when they do, and not before. (Used to have dozens “insider” twitter feeds that were very useful to me in my own wagering. Late starter sitting is a big deal, but those twitter feeds are now barely alive, so I’m actively seeking new reliable sources.)
Also, mainly because my stodgy bookie starts his week on Tuesdays, I’m doing the same here for convenience’s sake. Finally, my projections post includes a lot of data which can be useful and beneficial in making a bettor’s decision. Of course, no matter how relevant the data or elegant the algorithm, the nasty realities of context often win the day - but numbers do rule in the end.
GLTA
PS. Truth be told, I make a ton of mistakes, and sometimes bugs creep in – so, please, don’t hesitate to point out goof-ups.
Not a red flag, as such, is a poor ATS record; more important is the projected margin of victory (MOV) which decides everything. Posting, as I will, a few hours before tip-off means that late line movements and game-time starter-sitting, make a joke out of early projected ATS winners - but those are outliers, and nothing to do with the model’s reliability. Although I track the ATS record, far more important is variation in the MOV which, when it consistently goes wide, is a major red flag.
Totals projections have never been that reliable, so caveat bettor.
The model is based on past performance team and player data, rotations, injuries, and game matchups. The first posting of the week will include a standings frame for reference, and the thread will be updated with daily projection posts. Bear in mind, these are math-based projections, not “predictions”, and should be used only as a guide to betting; and, again, the posted MOV is far more useful to a betting decision than the ATS “pick”.
The final ATS pick will be based on the posted MOV and the consensus closing lines at VSiN which, for record keeping purposes may involve pick-flips, that late spread line adjustments may require. The posted MOV doesn’t change, but the final pick always depends on the closing line.
As I said, I’ll be posting hours before tip-off for the earliest games, except for early morning tip-offs on some weekends. Residing at 19.89° N, 155.66° W, I won’t be getting up in the dark to do this. The projection posts will come when they do, and not before. (Used to have dozens “insider” twitter feeds that were very useful to me in my own wagering. Late starter sitting is a big deal, but those twitter feeds are now barely alive, so I’m actively seeking new reliable sources.)
Also, mainly because my stodgy bookie starts his week on Tuesdays, I’m doing the same here for convenience’s sake. Finally, my projections post includes a lot of data which can be useful and beneficial in making a bettor’s decision. Of course, no matter how relevant the data or elegant the algorithm, the nasty realities of context often win the day - but numbers do rule in the end.
GLTA
PS. Truth be told, I make a ton of mistakes, and sometimes bugs creep in – so, please, don’t hesitate to point out goof-ups.
