No best bet for me tonight. BOL!
-
Comment -
NBA SDQL 'BEST BET' 2014-15
11-17-14
Plays - some maybe pending.....
Ronald S. - ORL +4 (1.05 to win 1)
Consigliere - PHX -2 (1.05 to win1)
Mako-SBR - NO OV 203 (1.1 to win 1)
nash13 - PHX -1.5 (1.09 to win 1)
Relay any errors please
bol fellas.. let's sweep 'em
Comment -
Forty Six...are you still part of the Best Bet group? Please let me know.Comment -
I'm tailing you guys on Phoenix. This thread is AWESOME.Comment -
definitely digging the card tonight. Good luck to us.Comment -
I think there is some value in college too, esp on the totals.Comment -
Overall Group Record - 20-13 60.6% (+6.35)
Yesterday's recap 3-1 (+1.90)
Official Entries: Max 10 of 10:
1. JMon - 2-3 (-.60)
2. pip2 - 2-1 (+.95)
3. nash13 - 4-1 (+3.00)
4. Ronald S. - 3-1 (+1.95)
5. hyahya - 2-2 (-.20)
6. Mako-SBR - 1-3 (-2.30)
7. FlyinAir - 1-2 (-1.45)
8. Consigliere - 4-0 (+4.00)
9. Sports Mind - 1-0 (+1.00)
10. Forty Six - 0-0 (0.00)
Comment -
0.96*(tA(FGA,N=3)+tA(turnovers,N=3)+0.44 *tA(FTA,N=3)-tA(offensive rebounds,N=3))>92 and 0.96*(oA(FGA,N=3)+oA(turnovers,N=3)+0.44 *oA(FTA,N=3)-oA(offensive rebounds,N=3))>92 and (tS(ou margin<0, N=5)>=3 or oS(ou margin<0,N=5)>=3) and total>=220 and HHere's one I got to thinking about last night....one thing SDQL can do, although it is a bit cumbersome, is basketball analytics. Some of the analytics applications I think is best in playing totals and this one is geared for that. In English.....two teams that play with above average # of possessions per game in their last 3 (ie above average pace), and one of the teams has gone under the total in at least 3 of their last 5 games bring the total down into a range that represents +EV, and the total posted is at 220 or above so the expected outcome is a high scoring game. This is a case where the books have to post a line people will see "value" in the under because of the high total but in reality both teams are primed to exceed the big number. I'm calling this the Phx Suns revolution query because it only applies for the last 6-8 years or so when Phx revolutionized the game. Not a huge sample size but looks pretty promising....1 1-0-0 (17.00, 100.0%) -9.0 0-1-0 (-27.00, 0.0%) 225.0 1-0 (26.00, 100.0%) season = 1995 2 1-1-0 (-3.00, 50.0%) -2.0 0-2-0 (-20.75, 0.0%) 221.8 1-1 (-1.00, 50.0%) season = 2012 9 5-4-0 (0.50, 55.6%) -1.6 7-2-0 (8.89, 77.8%) 225.9 5-4 (2.11, 55.6%) season = 2006 9 4-4-1 (-0.17, 50.0%) -2.4 6-3-0 (4.83, 66.7%) 221.8 5-4 (2.22, 55.6%) season = 2010 10 6-4-0 (3.15, 60.0%) -1.9 6-4-0 (5.80, 60.0%) 222.8 5-5 (5.00, 50.0%) season = 2004 11 4-7-0 (-2.73, 36.4%) -5.8 7-4-0 (8.23, 63.6%) 223.2 7-4 (3.09, 63.6%) season = 2013 21 13-8-0 (1.64, 61.9%) -1.8 14-7-0 (0.98, 66.7%) 227.3 12-9 (3.43, 57.1%) season = 2009 24 11-11-2 (2.31, 50.0%) -4.0 16-8-0 (9.27, 66.7%) 226.5 17-7 (6.29, 70.8%) season = 2008 27 10-17-0 (-0.94, 37.0%) -1.6 14-13-0 (0.89, 51.9%) 227.4 15-12 (0.70, 55.6%) season = 2007 Comment -
Queries to look at today:
den and no cover vs cle and sac from Ronald in sitpost #1389 ATS: 37-12-0 (3.71, 75.5%)
AD and p:AD and rest = 0 and o:rest > 0 and month = 11 and line < 4
ny covers vs mil from jmon sitpost #1392 -- ATS: 45-9-2 (7.15, 83.3%)
A and -3 <= line <= 5 and p:H and p:margin > 10 and P:W and month != 3 and o:rest < 2 and 2011 <= season and D
lal covers vs atl from jmon ATS: 97-49-2 (3.32, 66.4%)
A and tA(o
oints)>=103 and p:M2<=-10 and pp:M2<=-10
no covers vs sac from Ronald in sitpost #1389 -- ATS:58-24-0 (2.98, 70.7%)
AD and p:A and rest = 0 and o:rest>0 and month = 11 and line < 4Comment -
Pip, you posted the same query twice and like it was already pointed out, Denver was 12 points dog, not 1.5 points, so it's a mistake by KS.
Still, I like the logic behind this query a lot, but in my eyes, it needs to be adjusted. To either have a good team (to add high WP) or if it's a bad team, it needs to be off a loss, because the last thing you want, is to back a team like Jazz, Celtics, Magic and so on off a road win...
AD and p:AD and rest = 0 and o:rest > 0 and month = 11 and line < 4 and WP>50 18-6-0 (3.71, 75.0%)
AD and p:AD and rest = 0 and o:rest > 0 and month = 11 and line < 4 and WP<50 and p:L (12 - 3 - 80%)Comment -
And I have a question, is there is a way to use sdql, to calculate, at what point, is it better to take dog ML, than with the line?
We all had games, where we liked the dog to win, but wondered if it's best to take them ML or go safe and take the points.
I have zero doubt that there is a correct answer to that.
One guy that I respect a lot, told me once, that profit wise and in the long term, it's best to take dog ML, than the line, if it's up to 5 points. But still, would be nice to see numbers to prove it...Comment -
NBA SDQL 'BEST BET' 2014-15
11-18-14
Play New York Knicks +3 (-115)
AD and line <= 5 and p:H and p:margin > 10 and o:rest < 2 and 2011 <= season
Play an small away dog coming off a large home victory against a team with minimal rest. I also like the Knicks here as a contrarian given the preseason line for this game would have most likely been Knicks -1.Comment -
dmitean, at this point I am just quickly hunting down queries that pop up on my system and copying them over, but I did check that dupe out a bit. It was from the same post, but with a slight variation between p:A and p:AD, so I just stuck them both in there and let any viewers make their own choices (which they will do anyway).
Thanks for pointing out that DEN is a dud, that completely slipped by me.Comment -
NBA SDQL 'BEST BET' 2014-15
11-18-14 - 12:30p PST
Play: NO (+2.5) SAC -110
SDQL:
AD and rest=0 and o:rest>0 and line<4 and month=11 and season>=2006
SDQL English:
Playing the away dog on no rest against a rested home team, similar competition level due to the tight line, early in the season.
Notes: Believe this is one of JMon's early season scenarios, strong recently and over the long-term.
Also playing:
H and 60>= WP>=51 and tS(ou margin, N=5)>=18 and 198<=total<=207 and C and season>=2006
and
A and tA(ooints)>=103 and p:M2<=-10 and pp:M2<=-9 and o:rest>0 and game number>=8 and line<=13 and total<=232 and season>=2006
Must be a glutton for punishment on the first one, as it's burned me multiple times this season already after being a dominant scenario for so long, and that masochistic tendency continues into the second query by playing the Lakers, where you never know what you're going to get...Last edited by Mako-SBR; 11-18-14, 04:06 PM.Comment -
NBA SDQL 'BEST BET' 2014-15
11-18-14 - 4:22EST
Play: ATL vs LAL (UNDER 213.5) -105 @ Caribsports
SDQL:
pppppp:wins<=1 and opppppp:wins<=1 and conference=Eastern and o:rest<=1 and ats streak<=1 and total>195
SDQL English:
An early season trend that looks back at the win total both teams had 6 games ago, and if it was less than or equal to 1 for both then the game qualifies. Also, the team is in the Eastern conference and the opponent is 1 day rest or back to back, team's ATS streak is 1 or less and the total is greater than 195. Basically two teams that struggled to win in their first few games, game is expected to have a pretty good pace and some rest is factored in. Both teams have shown to be pretty variable on their scoring and this has the chance to be a blowout on the over, but I think that this goes under the very high total. Atlanta is very much like San Antonio, extra passes but can play with pace when required and LAL just stink. The LAL, SA game I think will be similar to this one which was in the 170s
Last edited by Consigliere; 11-18-14, 04:37 PM.Comment -
I like this premise, because it's gotten better over time and when you ask yourself why that would be, why would it be so much more dominant than say 10 years ago, one of the potential answers is the rise of the internet bettor. Books can't make the line in these games what it really should be, i.e. the favorite laying fewer points (or even getting points at home) because the square bettor likes betting home favorites, and too much money would drop on one side, leading to too much risk exposure.
As an example, in this case Detroit is apparently getting 52% of the money wagered as of this post time, which means the supposed 50/50 money balance books aim for has been achieved.
If they adjusted the line to what it should be, reflecting Orlando as a stronger team against Detroit (and thus getting fewer points), the 52% would climb as square internet bettors tend to like home favorites, and that would tip the money into a riskier proposition for the book as more money comes in on Detroit, like 60/40 or 70/30.
ATS streaks have previously been looked down on over the years, because they lead to Martingale-esque propositions that can destroy bankrolls through either chasing or fading the streak while doubling down into oblivion. But SDQL analysis can lead to insight into +EV where the book is backed into a corner, and this game and SDQL query is a possible example of that.
Good stuff Roland, even if it's a hypothesis that is way off base it's still interesting to see how it plays out.
Comment -
NBA SDQL 'BEST BET' 2014-15
11/18/14 5:00PM EST
New Orleans +2.5 -105 (5dimes) @Sacramento
AD and rest = 0 and o:rest > 0 and month = 11 and line < 4
SU: 83-62 (-0.01, 57.2%)
ATS: 95-48-2 (2.41, 66.4%) avg line: 2.4
O/U: 70-70-5 (0.30, 50.0%) avg total: 194.2
AD and rest = 0 and o:rest > 1 and month = 11 and line < 4
SU: 36-25 (0.39, 59.0%)
ATS: 43-18-0 (2.93, 70.5%) avg line: 2.5
O/U: 28-33-0 (-1.48, 45.9%) avg total: 193.6
Mako you read my mind! Not really much to say here as it's a trend we've used before and Mako explained it above. Going to play the short road dog on a b2b as it has been very profitable in the month of November (Teams are fresher early in the season). This is also a decent trend to fade around december-january. I included parameters for o:rest>1 (Kings are on 2 days rest) that show it's even better in that scenario.Comment -
I like both of those plays!Comment -
For those that are not looking at the spreadsheet, the 103 nba trends we are tracking daily there are a combined 31-16 so far on unique picks. The record does not count a pick where there are multiple trends supporting it (ie 3 trends, only 1 bet) and if there are contradicting trends the pick is cancelled. This is just over a week and a small sample size and assumes you play everything, which is not the intent, but a very promising start. There would be a nice way to have all the matchups highlighted automatically but at $2 a trend that would be a bit much! I've been tailing the sheet picks as 1/2 unit, combining those with the best bets makes for action packed nights and makes off for all the units being dumped on NFL!!! Keep up the good work fellas, this is gold!Comment -
Fingers crossed...after the lucky breaks at the end of the Phx/Bos game last night I feel like I am due for a letdown. Might be time for the fade...
selection=Best Bet and streak>=4 and pick=TOTAL
check that oneComment -
Same play where I took Bull ML...should have grabbed the line earlier but got busy with work..
NBA SDQL 'BEST BET' 2014-15
11-18-14 - 5:07 CT
Play Knicks +110
SDQL:
A and -3 <= line <= 5 and p:H and p:margin > 10 and P:W and o:rest < 2 and 2011 <= season and month!= 3 and playoffs=0
SDQL: English
Since 2011, road favorite of no more than 3 or dog of no more than 5, coming of a home win of more than 10 and won previous matchup v. oppt who is on no more than one days rest. As a dog role 74% ml winners with a margin of victory of 6.2Comment -
NBA SDQL 'BEST BET' 2014-15
11-18-14
JMon - NYK ML (1 to win 1.1)
hyahya - NYK +3 (1.15 to win 1)
Mako-SBR - NO +2.5 (1.1 to win 1)
Consigliere - ATL UN 213.5 (1.05 to win 1)
Ronald S. - NO +2.5 (1.05 to win 1)Comment -
Great work on this thread! The work here is fantastic. I'm just figuring out this SDQL world, but liking it so far.
Were Pip2's queries from this morning (Post #1832) just spitballing to get the day started? Two of them were plays anyway (NY and NO), but he also had the Lakers on a 97-49-2 ATS trend. Hindsight is easy, as it looks like they will cover...just wondering if it just slipped thru. Anytime I see a trend with 150+ plays over 60% to back it up, it gets my attention.Comment -
nice to see that the main purpose of the spreadsheet is getting value early. in mid week, i can not find enough time to post all the plays, but the beauty of working together is that there are others tracking, working on trends and putting things together.
i know there are some guys out there who could program a tool which could output trends automatically. there is an API tool on sportsdatabase but I am no expert on using it.Comment -
Results:Queries to look at today:
den and no cover vs cle and sac from Ronald in sitpost #1389 ATS: 37-12-0 (3.71, 75.5%)
AD and p:AD and rest = 0 and o:rest > 0 and month = 11 and line < 4
ny covers vs mil from jmon sitpost #1392 -- ATS: 45-9-2 (7.15, 83.3%)
A and -3 <= line <= 5 and p:H and p:margin > 10 and P:W and month != 3 and o:rest < 2 and 2011 <= season and D
lal covers vs atl from jmon ATS: 97-49-2 (3.32, 66.4%)
A and tA(ooints)>=103 and p:M2<=-10 and pp:M2<=-10
no covers vs sac from Ronald in sitpost #1389 -- ATS:58-24-0 (2.98, 70.7%)
AD and p:A and rest = 0 and o:rest>0 and month = 11 and line < 4
0-0: DEN covering vs CLE was a dud that I mistakenly copied from earlier games
1-0: Ronald's query correctly showed no covering vs sac
0-1: jmon's query missed on showing ny covering vs mil
1-0: jmon came right back with a query calling lal vs atl
1-0: Ronald's "tiny deviation" query also correctly showed no covering vs sac
Total: let's call it 2-1 for the evening, amidst the dud I mis-copied and the near duplicate query...Comment
...Geez, thanks did not notice...did you email Joe yet or que the group?? Probably messed with my automated queries.
