NFL - Situational Plays - SDQL

[emceeaye]

here are 2 intuitive offshoots of each other

Here are two I came up with that will hopefully come in handy at some point:

H and tS(COMP) / tS(passes) > .65 and tS(COMP) / tS(passes) > oS(COMP) / oS(passes) + .03 and tA(INT) < 1.25 and 100. * o:first downs / o:FDP < 25 and oS(passes) / oS(rushes) >= 1.15 and season >= 2005

and

H and tS(COMP) / tS(passes) > .65 and tS(COMP) / tS(passes) > oS(COMP) / oS(passes) + .03 and tA(INT) < 1.25 and o:FDP < 25 and oS(passes) / oS(rushes) >= 1.15 and season >= 2005

[FortySix]

WP=25 and p:L and op:W and 5<=t:week<=6 and -10<=t:line<=8.5 and (1989,1)<=(season,week)

42-15-2 ATS

WP=25 and p:L and op:W and DIV and 5<=t:week<=6 and -10<=t:line<=8.5 and (1989,1)<=(season,week)

18-1 ATS

Hey fellas,

These 2 are active tomorrow night with the 49ers @ Rams. Play the Rams. The first situation has had 4 straight ATS losses and the 2nd SDQL lost last week with the Saints. Whats your thoughts?

49ers are 9-0 ATS on MNF for a very long time but as well all know, trends do get broken. It is hard going against the 49ers but I think the Rams can do it. The Rams are 8-0 ATS when the team they are playing are on a 2+ win streak.

Personally I think the Rams can put up a good fight against the 49ers for MNF but we have all seen some crazy stuff on MNF this season so far. The overs might be a good play as well. I think the overs for prime-time games were hitting at something like 75% so far this season.

Appreciate some feedback. Thanks fellas.. Love this thread. Follow it all the time, with NFL and NBA. Just found the NHL one as well. Still learning all the formatting and SDQL language

[nash13]

Hey fellas,

These 2 are active tomorrow night with the 49ers @ Rams. Play the Rams. The first situation has had 4 straight ATS losses and the 2nd SDQL lost last week with the Saints. Whats your thoughts?

49ers are 9-0 ATS on MNF for a very long time but as well all know, trends do get broken. It is hard going against the 49ers but I think the Rams can do it. The Rams are 8-0 ATS when the team they are playing are on a 2+ win streak.

Personally I think the Rams can put up a good fight against the 49ers for MNF but we have all seen some crazy stuff on MNF this season so far. The overs might be a good play as well. I think the overs for prime-time games were hitting at something like 75% so far this season.

Appreciate some feedback. Thanks fellas.. Love this thread. Follow it all the time, with NFL and NBA. Just found the NHL one as well. Still learning all the formatting and SDQL language

hi there, this one is trending quite in the opposite way:
A and day=Monday and DIV and -8< line <6.5

[Mako-SBR]

hi there, this one is trending quite in the opposite way:
A and day=Monday and DIV and -8< line <6.5

Super strong the last few seasons, very nice.

[emceeaye]

Here is Dr. M's query of the day on the under:

team=Rams and HD and DIV and wins < o:wins and P: points>0 and season>=1996

The rationale of this query doesn't seem intuitively obvious to me. If someone sees it, can they explain? Thanks.

[emceeaye]

under query with good win % that can come in handy

Mistake

[Mako-SBR]

Here is Dr. M's query of the day on the under:

team=Rams and HD and DIV and wins < o:wins and P: points>0 and season>=1996

The rationale of this query doesn't seem intuitively obvious to me. If someone sees it, can they explain? Thanks.

The majority of his queries aren't based on logic or a real-world rationale, they're the result of him going into the DB and data-mining numbers down to ATS wins/losses...which is why he's lost on record more often than he's won on a seasonal yearly basis across all the sports he handicaps.

As I posted earlier, the hard part of SDQL handicapping isn't getting it to spit out ATS win rates of 80% over a large sample size via hundreds of queries. That's easy. The hard part is the human element, your own judgement in determining if the query situation is "real" (an actual advantage for any one team over any other, causation in other words) or "not real" just data-mined results that are just the result of random occurance lining up in the right way (flipping a coin 10x times heads does not mean the flipper knows how to produce a heads result, it's just a random run..correlation, not causation).

JMon is good at explaining it as well, but it's also the reason why you'll never see us parrot a Dr. M query here...because they lose, because they're not based on real-world advantages the majority of the time. (Note I said "majority" throughout this post, sometimes he does feature a query that is based on logical reasoning).

[Alex Vaile]

I've been doing quite well with Dr.M queries. Not bashing or hating Mako but hes been better than 50% each week. They are different from the queries most are putting on here but if u listen to the videos too they are very easy to follow. Tonite it's the under in the MNF game which I liked already even though most primetime games have been going over. NFC West games have been on trending under last few seasons and divisonal games in general tend to have less points scored. Just my opinion, best of luck to you all! He is 2 and 1 so far this week.

[Mako-SBR]

I've been doing quite well with Dr.M queries. Not bashing or hating Mako but hes been better than 50% each week. They are different from the queries most are putting on here but if u listen to the videos too they are very easy to follow. Tonite it's the under in the MNF game which I liked already even though most primetime games have been going over. NFC West games have been on trending under last few seasons and divisonal games in general tend to have less points scored. Just my opinion, best of luck to you all! He is 2 and 1 so far this week.

No worries, just be very careful and you'll be fine. His pick record is difficult to collect but easy to find at pregame. I did just that when i first learned about SDQL, analyzed his last 500 picks he sold to subscribers by hand, took a full week of work to do, and found the shockingly poor results over time (shocking to me at least, others know the score on him apparantly).

But I do feel that his work is good to use as a knowledge builder, particularly SDQL query building in terms of the parameters, but definitely NOT something you tail or bet on, ever.

Sounds like you're on the right track though Alex, this is more a warning for others who find SDQL, get intimidated by it, and then just say "oh I'll buy this Dr. M guy's picks instead of figuring out the tough parts".

hi there, this one is trending quite in the opposite way:
A and day=Monday and DIV and -8< line <6.5

Winner, good job Nash!

[emceeaye]

The majority of his queries aren't based on logic or a real-world rationale, they're the result of him going into the DB and data-mining numbers down to ATS wins/losses...which is why he's lost on record more often than he's won on a seasonal yearly basis across all the sports he handicaps.

As I posted earlier, the hard part of SDQL handicapping isn't getting it to spit out ATS win rates of 80% over a large sample size via hundreds of queries. That's easy. The hard part is the human element, your own judgement in determining if the query situation is "real" (an actual advantage for any one team over any other, causation in other words) or "not real" just data-mined results that are just the result of random occurance lining up in the right way (flipping a coin 10x times heads does not mean the flipper knows how to produce a heads result, it's just a random run..correlation, not causation).

JMon is good at explaining it as well, but it's also the reason why you'll never see us parrot a Dr. M query here...because they lose, because they're not based on real-world advantages the majority of the time. (Note I said "majority" throughout this post, sometimes he does feature a query that is based on logical reasoning).

Yes, I didn't wager on that game because I could not determine any intuitive rationale behind it. Now, as I stated before, that is not to say I don't go with the results of queries that aren't immediately obvious--If the sample size is large enough, I may make the wager even if I don't understand all of it right away. If the sample is 40-50 or less for example, with no rationale or one that is not intuitive, I will most likely not wager on it.

[nash13]

I calculated z-scores for Total and ATS trends, and take ROI, Number of games and last years profit into account.
After that I divide the trends into 3 groups of percentiles on each topic.
3 Points for the best percentile, 2 for the middle and 1 for the lowest.
At the end I calculate an average for each trend. I t starts at 1 (low performing) and finishes at 3 (best performing)
To calculate my wagers I ad pup the averages for the trend in the game and let them outweigh each other if necessary.
Highest wager so far 11%. Works fine:

https://docs.google.com/spreadsheets/d/1OzPC3WmCK6UgZVY-aLaoSkH3L9VjaAKa5Rtqky7OY4w/edit#gid=0

i backtested this variable staking for last year:
Profit of 1900 Units.
If anyone like to "Stresstest" his query, write it down or PM me.

[Mako-SBR]

Yes, I didn't wager on that game because I could not determine any intuitive rationale behind it. Now, as I stated before, that is not to say I don't go with the results of queries that aren't immediately obvious--If the sample size is large enough, I may make the wager even if I don't understand all of it right away. If the sample is 40-50 or less for example, with no rationale or one that is not intuitive, I will most likely not wager on it.

Yes, same for me as well, well said emcee. If the sample size is large enough and it's not something I consider to be volatile in general (like say team trends, which are much more likely to be obsolete or go bad more quickly than a time/travel/fatigue-general situation) even if I can't see the "why" in it, I'll still give it a test for a season to see how it does.

I calculated z-scores for Total and ATS trends, and take ROI, Number of games and last years profit into account. After that I divide the trends into 3 groups of percentiles on each topic: 3 Points for the best percentile, 2 for the middle and 1 for the lowest.

At the end I calculate an average for each trend. I t starts at 1 (low performing) and finishes at 3 (best performing)

To calculate my wagers I ad pup the averages for the trend in the game and let them outweigh each other if necessary.

Highest wager so far 11%. Works fine:

https://docs.google.com/spreadsheets/d/1OzPC3WmCK6UgZVY-aLaoSkH3L9VjaAKa5Rtqky7OY4w/edit#gid=0

i backtested this variable staking for last year: Profit of 1900 Units.

If anyone like to "Stresstest" his query, write it down or PM me.

Very nice Nash, that is great. I use a similar screening system based on z-score, z-score ratio, timespan, ATSm, etc, along with a variety of other tricks to try and protect myself from queries or systems that are just random and not predictive going forward.

It's not perfect but it does help.
Nice hits on Colorado and Washington btw, wow.

[Alex Vaile]

No worries, just be very careful and you'll be fine. His pick record is difficult to collect but easy to find at pregame. I did just that when i first learned about SDQL, analyzed his last 500 picks he sold to subscribers by hand, took a full week of work to do, and found the shockingly poor results over time (shocking to me at least, others know the score on him apparantly).

But I do feel that his work is good to use as a knowledge builder, particularly SDQL query building in terms of the parameters, but definitely NOT something you tail or bet on, ever.

Sounds like you're on the right track though Alex, this is more a warning for others who find SDQL, get intimidated by it, and then just say "oh I'll buy this Dr. M guy's picks instead of figuring out the tough parts".



Winner, good job Nash!

Right on thanks. Good insights. Keep it rolling. My niners burned me late on under but at least they won!

[JMon]

Play over two non-conference teams coming off ATS wins...small sample but worth a look! OV ARI/OAK

-4<=line<=4 and not C and p:ATSW and op:ATSW and 47>=total>=43 and season>=1997 and H

[emceeaye]

This one from Dr. M or perhaps someone else in this thread (forgive me if I'm not giving proper attribution) posted this:

Its on the Texans for tomorrow. It looks pretty good:

A and rest>4 and -3<=line<=3 and NDIV and p:L and WP=50 and NB and season>=2004

Thoughts?

[Alex Vaile]

This one from Dr. M or perhaps someone else in this thread (forgive me if I'm not giving proper attribution) posted this:

Its on the Texans for tomorrow. It looks pretty good:

A and rest>4 and -3<=line<=3 and NDIV and p:L and WP=50 and NB and season>=2004

Thoughts?

It looks great. They are more rested, just sucks as Pitt usually plays great on MNF. Dr. M also liking over tonite on there today. Im still undecided as Fitzpatrick is pretty terrible

[Mako-SBR]

It looks great. They are more rested, just sucks as Pitt usually plays great on MNF. Dr. M also liking over tonite on there today. Im still undecided as Fitzpatrick is pretty terrible

That query isn't bad from him, prefer when he stays away from either Team or Player trends...which are the most volatile and change (go obsolete) far quicker than the longer-lasting generic situation-based queries like that one.

Was on the Texans already, but my confidence is like yours Alex, nowhere near say the same level of comfort I had when placing the Denver/Baltimore/Dallas bets Saturday.

[Mako-SBR]

Play over two non-conference teams coming off ATS wins...small sample but worth a look! OV ARI/OAK

-4<=line<=4 and not C and p:ATSW and op:ATSW and 47>=total>=43 and season>=1997 and H

We should have won this one, had another query showing the same Over outcome. Next time we'll get it.

[Alex Vaile]

We should have won this one, had another query showing the same Over outcome. Next time we'll get it.

I played it too. Looked decent at the half. Raiders just stink

[JMon]

We should have won this one, had another query showing the same Over outcome. Next time we'll get it.

I had 4 queries pointing to the over, way it goes sometimes. Ya it did look good at half, alex

[JMon]

This play is on Den

SDQL fading SD. Simple logic..fading a road dog with a poor rushing team against a good rushing defense. Note parameter below. Also note dogs of >=8 and high totals of >=45

A and 10>=line>=3.5 and 95>=tA(rushing yards)>=70 and 95>=oA(o:rushing yards)>=70 and 2003<=season

Anyone have anything else???

[nash13]

I have some for the weekend, but waiting for lines to settle.

[Mako-SBR]

This play is on Den

SDQL fading SD. Simple logic..fading a road dog with a poor rushing team against a good rushing defense. Note parameter below. Also note dogs of >=8 and high totals of >=45

A and 10>=line>=3.5 and 95>=tA(rushing yards)>=70 and 95>=oA(o:rushing yards)>=70 and 2003<=season

Anyone have anything else???

Winner! Nice J, thanks for sharing it.

[JMon]

Boy this thread is crickets... not too many SDQL NFL fans I guess or not enough action, lol?!?

Add parameter looking at NFC

SDQL, fading the Panthers and over

season >= 2004 and H and 8 > line > 2.5 and rest < 6 and NCONF = NCONF and 16 > game number > 2 and total < 53 and conference=NFC

SDQL, suggesting NO

F and tA(o:PY)>=255 and po:YPPA>=7 and ppo:YPPA>=7 and pppo:YPPA>=7 and line>-7.5

[terrortwylight]

Boy this thread is crickets... not too many SDQL NFL fans I guess or not enough action, lol?!?

Add parameter looking at NFC

SDQL, fading the Panthers and over

season >= 2004 and H and 8 > line > 2.5 and rest < 6 and NCONF = NCONF and 16 > game number > 2 and total < 53 and conference=NFC

SDQL, suggesting NO

F and tA(o:PY)>=255 and po:YPPA>=7 and ppo:YPPA>=7 and pppo:YPPA>=7 and line>-7.5

JMon.. I'd love to understand SDQL but I just don't. Could you give me some insight on the basics so I can start using it when I handicap? Also, what does that even mean above? Play Saints and under?

[JMon]

JMon.. I'd love to understand SDQL but I just don't. Could you give me some insight on the basics so I can start using it when I handicap? Also, what does that even mean above? Play Saints and under?

The SDQL queries above point to NO ATS and over...but does not necessary suggest you should bet them. Using SDQL comes with years of practice and knowledge of how to use it. The above queries can be copied and pasted into the NFL query database over at killersports. For beginners, I would suggest downloading the manuals and get a grasp of the language of what makes a SDQL query. By googling SDQL and killersports you will find such material. Most of the individuals here are well-versed in the language and like myself, will help you along if you are interested. Unfortunately this SDQL thread is not active with SDQL posters, which is sad, but it appears the NHL is clicking and hopefully NBA will be soon! I was in the MLB thread daily and you will see me in the NBA thread a lot. Cheers.

[Mako-SBR]

SDQL, fading the Panthers and over

season >= 2004 and H and 8 > line > 2.5 and rest < 6 and NCONF = NCONF and 16 > game number > 2 and total < 53 and conference=NFC

SDQL, suggesting NO

F and tA(o:PY)>=255 and po:YPPA>=7 and ppo:YPPA>=7 and pppo:YPPA>=7 and line>-7.5

Two more excellent scenarios that despite having smaller sample sizes would be ones I'd play myself without hesitation. Nice as always J.

Unfortunately this SDQL thread is not active with SDQL posters, which is sad, but it appears the NHL is clicking and hopefully NBA will be soon! I was in the MLB thread daily and you will see me in the NBA thread a lot.

Same, just haven't been able to create any "special" (worth posting/playing) scenarios for the NFL this season due to not focusing on it enough time wise. NBA distracts me, just so much easier due to having so much data to work with.

[nash13]

I am very enthusiastic on NHL, the problem with NFL is the low number if games. It is not as productive as MLB or NBA.

[b1slickguy]

NCAAFB

Created this one a few seasons ago.
Small tracked sample, but you can see the lack of scoring prior to 2006.
Take the under with home teams after a bye week whose previous opposition was Alabama.
Good luck.

H and po:team=ALA and rest>12

[JMon]

Since 2008 in the month of November, look and see how home teams do after covering 5/6 ATS out of their last 7 and their oppt is on less than 9 days rest. Linesmakers are want Joe all over these favs. Look at the short fav >-3.5

SDQL:

4< tS(ats margin>0,N=7)< 7 and H and month=11 and 2008<=season and o:rest<9

[FortySix]

My first NFL SDQL. Its a play on Eagles ML vs the Texans

AF and op:AW and opp:AL and p:AL

[FortySix]

Since 2008 in the month of November, look and see how home teams do after covering 5/6 ATS out of their last 7 and their oppt is on less than 9 days rest. Linesmakers are want Joe all over these favs. Look at the short fav >-3.5

SDQL:

4< tS(ats margin>0,N=7)< 7 and H and month=11 and 2008<=season and o:rest<9

JMon are you going to play all 3? I find it hard putting money on the Jets

[emceeaye]

Since 2008 in the month of November, look and see how home teams do after covering 5/6 ATS out of their last 7 and their oppt is on less than 9 days rest. Linesmakers are want Joe all over these favs. Look at the short fav >-3.5

SDQL:

4< tS(ats margin>0,N=7)< 7 and H and month=11 and 2008<=season and o:rest<9

nice job Jmon!

[JMon]

JMon are you going to play all 3? I find it hard putting money on the Jets

nice job Jmon!

46- ya I played all three...think Jets had a shot without the fluke TD and Vick getting hurt.

Thanks EMC! Hope to see you in this thread more often bud!

[FortySix]

Hey fellas, I hope this thread is still going. Play for TNF

HF and points>32 and op:HW and p:HW and total >42 and opoints<24