0
0
Member
Okay, kids - I’m more than a month late to start, but until very lately the model’s projections have been too far off to even approximate reliability. I’ve waited to have a decent couple weeks behind a first posting, and it’s now or never. The trouble is easy to spot; teams just don’t play much defense this season, and the number of 3-point attempts is going through the roof. There’s also a kind of new species of “load management” that’s apparent; namely, starters are playing their normal minutes and shooting the same but playing less and less defense in the 3rd and 4th quarters. The model just isn’t structured to accommodate this kind of play and requires deep modifications to the algorithm, which I’m not ready to do now and will have to wait until the off-season. So….
Last couple weeks the model hit 52% ATS – whoohoo – and even the Over/Unders broke 50%, so I’ve decided to give the model a chance for at least one week. I’ve also decided to post my “Bests Bets” each day – the games I decide to wager on – up to 4 games that look good to me, but more likely fewer. I’ll post these along with the full slate of games, as close to the first game tip-off each day (unless I’m busy, or the first game starts while I’m still in bed, which is a Hawaii thing). Close to tip-off, mainly because of the increase numbers of GTDs you see at this point in the season. Which is to say, there may be days I don’t play any of my “Best Bets” at all, and I’ll try to post “No Play” just prior to that game’s tip-off, but no promises. I’ve been breaking slightly better than even so far, so there might be value in posting.
Again, please don’t hesitate to tell me if I screw up somewhere. I do a lot of work on this stuff each day, and mistakes come with the territory.
Once again, I’ll post daily, each week beginning on Mondays until, and unless, the model starts tanking, which it tends to do. I’ll post my updated standings (not easy to do given the number of postponements) in the first post each week; and remember, the model’s daily updated ATS record Is based on the consensus closing lines according VsIN, although my “Best Bets” uses the lines my house provides.
Other Notes: The basic model uses past performance of teams and players, along with matchups and injuries, to produce its projections; however, my “Best Plays” – the plays I like - while leaning on those projections, rely more on perceived momentum, and will often turn the model’s projected ATS winner on its head. Similarly, I’m more sensitive to line changes than the model which, again, could result in a difference of opinion (I guess, with myself), so don’t be surprised. Finally, because of the numerous postponements, the Last Ten Games win/loss data is pretty screwed up – it requires a modification that will have to wait – but I include it because I use it anyway.
The first week or so there are bound to be glitches, and if you spot one, please let me know. GLTA
Last couple weeks the model hit 52% ATS – whoohoo – and even the Over/Unders broke 50%, so I’ve decided to give the model a chance for at least one week. I’ve also decided to post my “Bests Bets” each day – the games I decide to wager on – up to 4 games that look good to me, but more likely fewer. I’ll post these along with the full slate of games, as close to the first game tip-off each day (unless I’m busy, or the first game starts while I’m still in bed, which is a Hawaii thing). Close to tip-off, mainly because of the increase numbers of GTDs you see at this point in the season. Which is to say, there may be days I don’t play any of my “Best Bets” at all, and I’ll try to post “No Play” just prior to that game’s tip-off, but no promises. I’ve been breaking slightly better than even so far, so there might be value in posting.
Again, please don’t hesitate to tell me if I screw up somewhere. I do a lot of work on this stuff each day, and mistakes come with the territory.
Once again, I’ll post daily, each week beginning on Mondays until, and unless, the model starts tanking, which it tends to do. I’ll post my updated standings (not easy to do given the number of postponements) in the first post each week; and remember, the model’s daily updated ATS record Is based on the consensus closing lines according VsIN, although my “Best Bets” uses the lines my house provides.
Other Notes: The basic model uses past performance of teams and players, along with matchups and injuries, to produce its projections; however, my “Best Plays” – the plays I like - while leaning on those projections, rely more on perceived momentum, and will often turn the model’s projected ATS winner on its head. Similarly, I’m more sensitive to line changes than the model which, again, could result in a difference of opinion (I guess, with myself), so don’t be surprised. Finally, because of the numerous postponements, the Last Ten Games win/loss data is pretty screwed up – it requires a modification that will have to wait – but I include it because I use it anyway.
The first week or so there are bound to be glitches, and if you spot one, please let me know. GLTA