NBA Stituational Bet, SDQL

[nash13]

Here is one of my own. I call it "Sunday letdown"
HD and day = Sunday and rest < 3 and o:rest < 3 and total > 185 and n:H and season>=2006

[dmitean]

I think you are right, except maybe "month" is just our best working approximation of clumps of time where teams have slightly different outlooks on b-ball and b-ball in relation to their lives in general. The kinds of markers the team personnel might be paying attention to might be Xmas, Thanksgiving, all-star break, and playoffs, maybe, more than months...

I agree, that's why months matter though. I believe that team in biggest fatigue in January. Start of the season is far behind, the holidays are behind and ASG break no where in sight. Favorites tend to overlook the most times in this month I believe and bad teams many times just don't show up. In February is break or make time for many teams. I know that Toronto for example, always been really focused to go out on the break on a high note. Many teams try to improve their positions at that time - giving everything they got, because they know that the break is near and they will have time to rest and want to go with a good taste.
Also, it's close to trade deadline, so many teams check if they got what it takes to make one last push and if the answer is no, they can break their team and shop half of it.
End of March is where the tanking begins and fight for the playoffs really heats up - sometimes it can even start earlier.
April - time to tank, for good teams to get some rest for their key players, before playoffs, for some teams, final chance to make the playoffs.
November is the month where the fatigue still haven't kicked in, teams still full of hope, but new teams, still have some chemistry issues.
I believe that each month is different and that it's possible for trend not to work in some month.

[pip2]

I agree, that's why months matter though. I believe that team in biggest fatigue in January. Start of the season is far behind, the holidays are behind and ASG break no where in sight. Favorites tend to overlook the most times in this month I believe and bad teams many times just don't show up. In February is break or make time for many teams. I know that Toronto for example, always been really focused to go out on the break on a high note. Many teams try to improve their positions at that time - giving everything they got, because they know that the break is near and they will have time to rest and want to go with a good taste.
Also, it's close to trade deadline, so many teams check if they got what it takes to make one last push and if the answer is no, they can break their team and shop half of it.
End of March is where the tanking begins and fight for the playoffs really heats up - sometimes it can even start earlier.
April - time to tank, for good teams to get some rest for their key players, before playoffs, for some teams, final chance to make the playoffs.
November is the month where the fatigue still haven't kicked in, teams still full of hope, but new teams, still have some chemistry issues.
I believe that each month is different and that it's possible for trend not to work in some month.

Yep, great breakdown of the mini-parts of a season. Maybe we should try to define them with X>game number>Y types of ranges...

[Cutler'sThumb]

I don't see any reason why ST trend is weaker than LT trend, for short term...

If there is a trend that works this season, that's all I care about. If trend was great for 10 years, but doesn't work this season, I won't use it.
I always check many things, but 2 main ones - how the trend is doing this season and how it does in the current month overall (previous seasons as well of course - but current month).
There are trends that are great, but in one month, for some reason just don't work. Don't forget that each month in the NBA is different. Fatigue wise, importance wise, rest wise.
Each month stands for something else, though December and January are similar imo more or less.

I was thinking of this in context of there being 4 other trends supporting OKC, and the one ST trend supporting Miami. If someone were eliminating any conflicting plays, this one would have been out due to the one relatively weak conflicting trend. Somewhat similar to Nash13's "only rated above 3" back test.

[nash13]

i eliminate every play even if the conflicting one is the weakest i have.

[nash13]

74 query also supports the Thunder.

Reported the missing queries bug to the developer team. They fixed it in yesterday. Seemed to be a problem with API on Killersports. Now it is working just fine. Fast support.

[nash13]

H and (points + opoints) / 2 > 100 and line > 4 and rest < 3 and WP > 15 and o:WP < 75 and season>=2006

[nash13]

Been working on new angles. I call it reverse strategy. Since 70% og the plays the queries generate are based on Away Teams, the rest left out should be Home ATS wins. I will back test them later.

[nash13]

Added NBA plays of this season up to date to the sheet.

[nash13]

Solid 13% Yield and 168 Units Profit this season so far.

[nash13]

Backtested "Reverse Psychology" 58-45 so far this season. +12.38 Units on single bets.
Together with the single plays from the queries:
451-336
approx 80 units

[figue]

Backtested "Reverse Psychology" 58-45 so far this season. +12.38 Units on single bets.
Together with the single plays from the queries:
451-336
approx 80 units

how is the reverse psychology ??

[nash13]

Reverse psychology means. I assume that H and A on ATS will even out in the end.
Since most of the trends are predicting away teams to win, i will take the games left out for the home team to win ats.
Let's say no queries show up for a game, i take the Home Team ATS.
58-45 on singles this year.

Only singles:
Queries for Away Teams so far this season:
167-110 +41,97
Queries for Home Teams so far this season:
40-46 -7,67

So you can see the system is very good at predicting away winners on ats.
but there have to be home winners too? these are the games left out.

[figue]

Reverse psychology means. I assume that H and A on ATS will even out in the end.
Since most of the trends are predicting away teams to win, i will take the games left out for the home team to win ats.
Let's say no queries show up for a game, i take the Home Team ATS.
58-45 on singles this year.

Only singles:
Queries for Away Teams so far this season:
167-110 +41,97
Queries for Home Teams so far this season:
40-46 -7,67

So you can see the system is very good at predicting away winners on ats.
but there have to be home winners too? these are the games left out.

gracias

[dmitean]

I think you need to be an action junky to play games that showing up on queries and look for meaning, why games didn't show up on queries to bet the opposite.

I do wonder, why queries are so good picking away teams and much worse at picking home teams? I can add that AF seems to be team's best position for the queries...

[nash13]

I think you need to be an action junky to play games that showing up on queries and look for meaning, why games didn't show up on queries to bet the opposite.

I do wonder, why queries are so good picking away teams and much worse at picking home teams? I can add that AF seems to be team's best position for the queries...

Hmm it is just another angle to work with the queries. I guess the lines for Away Teams esp Favs are underestimated by the bookies under certain circumstances. The queries are pointing them out mostly.

[JMon]

showcasing a group-by btwn two teams on an over streak, tends to go under.


ou streak>=8,7,6,5,4,3,2 and ou streak>=9,8,7,6,5,4,3,2 and 2009<=season

[Mako-SBR]

Saw one on Statfox today that I thought was interesting, but can't get the SDQL numbers to match their historical records:

Play On - Road teams (SACRAMENTO) revenging a blowout loss vs opponent of 20 points or more against opponent off a home win by 10 points or more. 46-18 over the last 5 seasons. ( 71.9% | 26.2 units ) 1-2 this year. ( 33.3% | -1.2 units

When I tried to replicate the scenario for their 'last 5 season' result I used this query:

AD and P:L and P:season=season and P:margin<=-20 and op:HW and op:margin>=10 and 2013>=season>=2009

That produces the right number for this season (1-2 ATS), which matches what Statfox quotes, but doesn't come close to matching what Statfox shows for the scenario's last 5 seasons (2009-2013)...

[JMon]

^...Within the situation I'm not seeing where they are call for a road dog nor a same-season revenge game?

[Mako-SBR]

^...Within the situation I'm not seeing where they are call for a road dog nor a same-season revenge game?

Yeah I added the same season after getting the "1-2" record in 2014, meaning without it there were too many games already played in 2014 to match Statfox's published record so far. Once I did, it matched, and the 5-year totals are close also, but ATS wins and losses aren't, outside of the correct 1-2 so far this season. They were all dogs, hence just adding the D, doesn't change anything.

[JMon]

Yeah I added the same season after getting the "1-2" record in 2014, meaning without it there were too many games already played in 2014 to match Statfox's published record so far. Once I did, it matched, and the 5-year totals are close also, but ATS wins and losses aren't, outside of the correct 1-2 so far this season. They were all dogs, hence just adding the D, doesn't change anything.

Not sure then.. but I can tell you that statfox rounds up and the DB we use doesn't! So any numerical value there is going to be a difference between the two.

[Mako-SBR]

Not sure then.. but I can tell you that statfox rounds up and the DB we use doesn't! So any numerical value there is going to be a difference between the two.

Ah that's right, forgot about that! Thanks J.

[jeffjam_]

Hi, I am looking for a help of someone who can work with SDQL, think killersports.com. My basic idea that could possibly (or not) turn into a small system is that bad NBA teams are giving up as the season is coming to the end so the home teams should't be favoured by more than say 6 points because home court advantage disappears. There is no easy way to backtrack it other than using SDQL language, which I might learn, but given the fact that I am busy as hell, it is not possible at the moment. So if anyone able to make a querry in SDQL language would be of help, I would be really grateful. Parameters that I look for are as follows:

- games were played between 1st of March and the end of the season.
- both teams had under 40% straight up winning record, would be good to sequence it say 45%, 40%, 35% to see if results are different.
- home team is favoured by more than 6 points, again sequencing would be good, say 6+, 7+, 8+, 9+, 10+ favourite.
- so to sum it up, looking for ATS results of away teams with a winning record of 40% or less, catching (being a dog) 6 and more points against home team with winning record also 40% or less and games were played beyond 1st of March.

I would like to go as back as 5 years at the most.


Thank you for the help.

[nash13]

Bulls	Heat
Cavaliers	Thunder	Thunder +3.5 1u	Over 212 1u
Suns	Clippers
Pelicans	Mavericks	Mavericks -3 4u	Under 199.5 1u
Hawks	Timberwolves
Magic	Pacers
Spurs	Bucks	Spurs -11.5 1u	Under 190.5 1u
Raptors	Pistons	Pistons +8 2u	Over 200.5 1u
Warriors	Celtics	Warriors -17.5 1u	Over 214 1u
Nuggets	Wizards
Lakers	Rockets		Under 207 1u

This would be my plays tonight

[emceeaye]

1st quarter Clippers ATS

play against a road team, dog, and/or road dog with no rest ATS for the 1st quarter if they had the lead at the end of the 3rd quarter of yesterday's game but nevertheless, ended up losing the game.

2>=p:M3>0 and AD and 0>p:margin>=-2 and line>8 and rest = 0
2>=p:M3>0 and AD and 0>p:margin>=-2 and line>8 and rest = 0 and [M1>-4]

seems to work better when the final points differential was less than around 3.


It works nicely even though Clippers also on a back-to-back and had to travel coming into tonight's game

7>p:M3 >=0 and (AD) and 0> p:margin>=-7 and rest = 0 and o:rest=0 and A and op:A and line>10

[Alex Vaile]

play against a road team, dog, and/or road dog with no rest ATS for the 1st quarter if they had the lead at the end of the 3rd quarter of yesterday's game but nevertheless, ended up losing the game.

2>=p:M3>0 and AD and 0>p:margin>=-2 and line>8 and rest = 0
2>=p:M3>0 and AD and 0>p:margin>=-2 and line>8 and rest = 0 and [M1>-4]

seems to work better when the final points differential was less than around 3.


It works nicely even though Clippers also on a back-to-back and had to travel

7>p:M3 >=0 and (AD) and 0> p:margin>=-7 and rest = 0 and o:rest=0 and A and op:A and line>10

I like it too as the nuggets game went to ot also.

[hyahya]

Here's one from Statfox:

-3<=Average(margin@team and season) <= 3 and p:margin>15 and -3<=oA(margin@team and season)<=3 and HF and season>2009

[nash13]

where do you get this stat fox trends?

[hyahya]

where do you get this stat fox trends?

I usually go to the RX service plays forum and select their daily service plays thread; Statfox trends are on the first page of the thread.

[nash13]

will add a sheet for the stat fox sheet trend which can not transferred automatically to the sheet. 1st half ATS, ML etc

[nash13]

H and total >= 200 and p:AL and p:margin <= -10 and day = Sunday
HD and p:FL and 60<=WP<=75 and 60<=o:WP<=75
HF and line <=-10 and p:H and points>=100 and pooints>=100 and t:date - p7:date = 13

[nash13]

total >= 200 and tA(rebounds-o:rebounds) > 3 and day = Tuesday and season >=2009

[nash13]

total >= 200 and tS(ats margin>0,N=4)=3 and o:WP>50 and WP>=75

[nash13]

Can someone translate this one? Seems like i am doing something wrong:
Play On - Road favorites of 3.5 to 9.5 points good offensive team - scoring 99+ points/game on the season, after allowing 55 points or more in the first half in 2 straight games

I have:
AF and line -3.5<=line<=-9.5 and tA(points@season)>=99 and po:P1+P2>=55 and ppo:P1+P2>=55

[emceeaye]

Can someone translate this one? Seems like i am doing something wrong:
Play On - Road favorites of 3.5 to 9.5 points good offensive team - scoring 99+ points/game on the season, after allowing 55 points or more in the first half in 2 straight games

I have:
AF and line -3.5<=line<=-9.5 and tA(points@season)>=99 and po:P1+P2>=55 and ppo:P1+P2>=55

this:
AF and -3.5>=line>=-9.5 and tA(points@team and season)>=99 and (po:P1+po:P2)>=55 and (ppo:P1+ppo:P2)>=55