East is a tool, I said it first
-
Comment -
As you research SDQL for the first time you inevitably come across StatFox's scenarios if you have even half a brain.
Once you do, you go through all of their various situations while using your head to filter out the ones that don't have a logical basis and are just data-mined versus keeping the ones you do believe in.
For East to get all crazy, literally crazy, about StatFox scenarios as all being "worthless" made no sense to me. Who gives a shit if the line is off a half point, etc?
Some of the best NBA performers I had this season were generated from SF's daily shout-out page...hilarious looking back on that discussion you guys had now and East going apeshit over nothing...his loss.
Comment -
is nothing more than a hypocrite. Hypocrisy and his "so called" methodology can be searched via the google group with ease. I don't like liar's and those claiming to be someone they are not.As you research SDQL for the first time you inevitably come across StatFox's scenarios if you have even half a brain.
Once you do, you go through all of their various situations while using your head to filter out the ones that don't have a logical basis and are just data-mined versus keeping the ones you do believe in.
For East to get all crazy, literally crazy, about StatFox scenarios as all being "worthless" made no sense to me. Who gives a shit if the line is off a half point, etc?
Some of the best NBA performers I had this season were generated from SF's daily shout-out page...hilarious looking back on that discussion you guys had now and East going apeshit over nothing...his loss.Last edited by JMon; 04-25-14, 06:59 AM.Comment -
A and playoffs = 1 and -3 <= line <= 0 and M2 > 2
SU: 35-6 (10.29, 85.4%)
ATS: 32-7-2 (8.68, 82.1%) avg line: -1.6
O/U: 15-24-2 (-0.80, 38.5%) avg total: 190.9Comment -
Nice fine Pip...have to save that one.
I am throwing this next stuff out for discussion, and there could/should some mistakes on my assuming certain things that will be apparent below. There was some talk about half time betting and third quarter betting earlier in the thread. I've never really done much of it, but perhaps someone will find something valuable to them, and maybe even to the group as a whole.
I mentioned that the half time method was to take an expected result at half time and to bet the second half with the expectation of what happened in the first half to continue. I'll discuss and display how this worked out so far this year with totals. The parameters of my discussion will be on totals only, as I believe (assume) that the total of the second half total line will be half of what the first half total is. With sides there is greater discrepancy according to how the first half has resulted, and I don't wish to put the effort into finding the second half betting lines....I'll leave that up to someone else.
The parameters are to bet the second half if the first half result is either >=6 points too high or too low from the expected first half result on the total. If the total for the game is 200, we'd expect the total to be 100 at the half.
If the first half result is less than 94 we'd bet the under in the second half, and if the first half result is greater than 106 we'd bet the over in the second half if the total was 100.
Listing the relevant games:
combined 1st half combined 2nd half
Date Total opp team points scored bet points scored result
4/19 186 Pacers Hawks 100 OVER 94 W
4/20 203 Mavs Spurs 87 UNDER 88 W
4/20 177' Wizards Bulls 102 OVER 93 W
4/20 216 T'Blazers Rockets 97 UNDER 145 L
4/22 187 Hawks Pacers 100 OVER 86 L
4/22 190' Nets Raptors 84 UNDER 111 L
4/22 180 Wizards Bulls 105 OVER 95 W
4/23 187' Bobcats Heat 104 OVER 94 W/T
4/24 187 Pacers Hawks 77 UNDER 106 L
4/24 190 Thunder Grizzlies 80 UNDER 113 L
4/24 213 Clippers Warriors 89 UNDER 105 WComment -
Comment -
I remember doing that when I first started; thinking I found the holy grail of queries, come to find out a game would never come active...Careful, you forgot the p : before that M2 parameter, which means you're not getting results from previous games moving forward. Happens to the best of us.
Corrected results:
29-24 (-0.25, 54.7%) 27-25-1 (-1.85, 51.9%) avg line: -1.6 20-33-0 (-3.94, 37.7%) avg total: 192.7
Comment -
Uh oh! Can you guys tell me the nature of the results of what I am actually getting by leaving out the p? My intent was that the game was actually at half time last night, and I didn't want to see the results of when the margin of the previous game's half time fell in a certain range. I wanted the results that happen in the same game as when the margin at halftime is greater than or lesser than a certain number...Comment -
You are correctly querying what you are looking for, Pip. I have used a few of these for live action. Since the last filter is not predictive you have to either create the query on the fly or just know your own saved queries well enough to spot the situations while the game is in progress and act accordingly. I like your outside the box thinking.
Good luck.Comment -
Your appraoch seems correct, but one thing I would improve is that margin range. Because now you are only filtering margins greater than 2 without any limitation, but the thing is that it is a big difference wether margin is 3 or, for example, 15, if opening line is around -1 to -3.Comment -
Ah, OK, good! I started going through the recaps of the listed games to make sure they fit the criteria, and they all seemed to fit my intent. But for all I know, there were a bunch left out or something.
In going over the recaps I saw that for a lot of those games teams had double digit leads at half time, so duh, yes they were very likely to beat the spread at the end of the game. But a double digit lead hardly applied to the 4-point lead the clippers had last night. So I changed the query to
A and playoffs = 1 and -3 <= line <= 0 and 8 > M2 > 2
SU: 15-3 (5.50, 83.3%)
ATS: 12-4-2 (3.94, 75.0%) avg line: -1.6
O/U: 4-12-2 (-5.53, 25.0%) avg total: 189.5
And now the sample size isn't so great...Comment -
p:M2>4 and AD and playoffs=1 and p:HFW
Big playoff sample.Last edited by green7; 04-25-14, 11:54 AM.Comment -
p:M2<0 and AD and playoffs=1 and p:HFLComment -
p:M2>=3 and AF and playoffs=1 and p:HWComment -
I tried tacking on a "series game>1" onto this to avoid the p: referring to the last game of the regular season. That seems to gain another 5%, if I am looking at it right...
p:M2 < 0 and AD and playoffs = 1 and p:HFL and series game > 1
SU: 17-31 (-4.27, 35.4%)
ATS: 23-25-0 (-0.11, 47.9%) avg line: 4.2
O/U: 14-34-0 (-5.93, 29.2%) avg total: 190.4Comment -
Good find Pip,...love it when someone can improve upon something that looks promising.
playoffs = 1 and HD and round = 1 and p:TPP < 35 and series game = 4
playoffs = 1 and round = 1 and p:TPP < 35 and AD and p:HW and series game<6 and po:TPP<35
ppp:HL and pp:HW and p:A and A and playoffs = 1 and round = 1
po
oints-points>15 and playoffs = 1 and round=1 and p:HF
Last edited by green7; 04-25-14, 11:59 AM.Comment -
JMon & Mako,
I love reading idiots' posts that are going nuts over nothing!
Where did this discussion take place?
TIA!Comment -
Nice pip and green, couple of solid winners in your posts, especially with a few tweaks.
It happened a while ago, last year if memory serves, on the Google Group. I found it prior to starting this thread while exhausting every possible search resource towards finding pre-existing NBA-based SDQL queries that were already out in the wild.Comment -
Careful with the previous query.....
p:M2 < 0 and AD and playoffs = 1 and p:HFL and series game > 1
If you add "and pp:HL" the UNDER results are not so great, though very small sample size. The results off a team winning the first of two home games and losing the second is spectacular, which isn't in play tonight.
The tweaked small query favors the Bulls and the Rockets...I'm watching (not playing the game) to see if the Rockets' stars actually show up this series....which to me if the most interesting series of the playoffs. Aldridge dominating everything and everyone.....best performance of the post-season so far.Comment -
I will try again tonight
total >= 200 and playoffs = 1 and series game = 3 and 2003<=season and rest<3 and HComment -
p:FL and playoffs=1 and series game>1 and t:series wins>o:series wins and HF
versus Blazers
p:M2 < 3 and playoffs = 1 and round = 1 and HF and p:L
favors Blazers
p:M2 < 3 and playoffs = 1 and round = 1 and AF and p:L
indication for 3 unders
p:TPP>40 and p:AL and A and playoffs=1 and total>195
Spurs
Comment -
I'm pasting this to try to describe a difficulty I am having with the playoffs:
playoffs = 1 and p:AW and pp:AW and series game > 2
SU: 12-5 (5.47, 70.6%)
ATS: 8-9-0 (-1.44, 47.1%) avg line: -6.9
O/U: 10-7-0 (0.24, 58.8%) avg total: 191.9
To me that query is a simple start to describing the basics of the situation between the Blazers and Rockets last night: one team won both of the opponent's first 2 home games of the series. But the problem for me is there's not much room left after this for trying to define the situation further query-wise, because I'm already down to a sample size of 17 games. Any other parameters I add in will take the sample size down to 2 or 3...
It feels like the query language is easier to use on regular season games than playoff games...Comment -
Can someone assist with including the proper query language to express the following?
games in which: the total first quarter points are at least 20% above the average for 1st quarter points; the total 2nd quarter points are at least 20% above the average for 2nd quarter points; and the total 3rd quarter points are at or below average for 3rd quarter points.
This is what I have but I don't think its working the way I want because the results show the total 1st quarter points are not at least 20% above the average of 47 1st quarter points as generated by green's query earlier.
Average(P1@team and season)>=1.20*Average(P1@season) and Average(P2@team and season)>=1.20*Average(P2@season) and Average(P3@team and season)<=Average(P3@season)
Thanks for any guidance in advance.Comment -
OK as a total rookie, not likely I am getting this correct, but here is my stab:
quarter scores P1 >= 1.20 * Average(P1@team and season) and quarter scores P2 >= 1.20 * Average(P2@team and season) and quarter scores P3 <= Average(P3@team and season)
Or the way I wrote it, before killer sports translated it:
P1>=1.20*Average(P1@team and season) and P2>=1.20*Average(P2@team and season) and P3<=Average(P3@team and season)
SU: 607-274 (6.49, 68.9%)
ATS: 579-280-22 (5.18, 67.4%) avg line: -1.3
O/U: 707-160-14 (12.23, 81.5%) avg total: 196.1Last edited by pip2; 04-26-14, 02:34 PM.Comment -
Yes, agree, which reminds me that JMon said at the start of the NBA post-season that he doesn't get heavy into it for that exact reason.
As an example for others, using the same processes and filtering and analyzing that led to a 77-48 record for the NBA regular season for my inaugural first-time effort with SDQL by using queries in this thread, queries from around the web, and queries that were self-generated, I've isolated 17 similar queries that apply to the post-season exclusively.
Last season those 17 queries indicate they produced a 52-8 record in the NBA post-season.
This season, my first NBA post-season using SDQL, they're now at 5-8...
Now it has been a crazy post-season so far, flipping the script on the NBA which usually has home favs dominating the spread the first 3 rounds, so that has to be factored in to the poor results.
But the reality is that the 17 queries simply don't have a large enough sample size to determine if they're overfit or actually valid prior to playing them, which makes it harder to disqualify the purely random scenarios from the consistent winners without actual wagers.
It's the same problem in the NFL regular season, where the super-small 16 game sample sizes combined with dramatic rule changes each year lead to it being very tough to handicap using pure SDQL plays. MLB and NBA regular seasons are infinitely easier due to the larger amount of games played, you can really determine which queries are nonsense (data-mined) just by looking at them and a few metrics thanks to the big numbers.
Still fun to try and solve the riddle, and I'm going to keep playing the 17 NBA post-season queries at smaller unit sizes just to get data for next season, but I would definitely advise caution for those lurking who came into the thread later in the year.
Comment -
Again not a lot of sample size here, but kind of an indicator, if you ever happen to notice Aldridge at the half is on track for a sub 40 shooting-percentage game:
Trailblazers:LaMarcus Aldridge:field goals made / Trailblazers:LaMarcus Aldridge:field goals attempted < .40 and playoffs = 1
SU: 1-5 (-14.50, 16.7%)
ATS: 0-6-0 (-15.00, 0.0%) avg line: -0.5
O/U: 3-3-0 (-1.50, 50.0%) avg total: 199.3Comment -
Trying this as an air bet at halftime of a DAL/SA:
playoffs = 1 and -3 > M2 > -7 and -1 < line < 3
SU: 7-19 (-6.31, 26.9%)
ATS: 7-18-1 (-4.69, 28.0%) avg line: 1.6
O/U: 9-16-1 (-1.54, 36.0%) avg total: 192.2
The second half line was +4 for DAL, with DAL up by 5, effectively moving the game line to +1 for Spurs from the original -3 Spurs. So this query should show results for a playoff team (Spurs) trailing by 2-6 points at halftime, with a +1 line.
Not sure if I can integrate the 2nd half spread with the game spread without messing everything up.....
...pushLast edited by pip2; 04-26-14, 07:06 PM.Comment -
Thanks, Mako. I have used Statfox and sometimes their stuff is good, many times needs a little massaging, and sometimes totally incorrect. But, if you talk to the Statfox people, they usually disagree.Nice pip and green, couple of solid winners in your posts, especially with a few tweaks.
It happened a while ago, last year if memory serves, on the Google Group. I found it prior to starting this thread while exhausting every possible search resource towards finding pre-existing NBA-based SDQL queries that were already out in the wild.Comment -
OKC/MEM halftime: 2nd half odds are pk, and OKC leads by 7, effectively making a new game spread of -7 for OKC:
playoffs = 1 and 9 > M2 > 5 and -9 < line < -5
SU: 50-8 (8.72, 86.2%)
ATS: 33-24-1 (1.84, 57.9%) avg line: -6.9
O/U: 30-28-0 (-1.05, 51.7%) avg total: 189.8
....lossLast edited by pip2; 04-26-14, 11:43 PM.Comment -
anyone have something on the Clips????Comment -
anyone have something on the Clips????
-3<=line<=3 and P:L and P:margin>=-3 and p:L and p:margin>=-3 and p:division=po:division
H and p: points<100 and pp: points<100 and tA(points)>=100 and 1.5<=line<=3 (saved from someone else)Comment -
Good luck JMon,....still have not pulled the trigger
HD and playoffs = 1 and series game = 4 and p:TPP < 35 and po:TPP < 35 and round = 1 and p:LComment
