MLB - Situation Plays - SDQL

[BasesLoaded314]

month = 7 and F and season > 2007 and -149 > line > -190 and s:SRA <
6 and ss:SRA < 7 and 2.5 > STDSWHIP > 1.1 and day != Monday and day !=
Tuesday and conference = o:conference and (p:hits + pp:hits) < 28 and WP <
59 and (starter wins - starter losses + o:starter losses - o:starter wins) >
-2

check out how it performs on Wednesdays

[BasesLoaded314]

H and -8 > (starter wins - starter losses + o:starter losses - o:starter wins) > -14 and day != Wednesday and day != Thursday and day != Friday and 6 > s:SRA > 0 and p:hits < 13 and total > 6.5 and (1.4 > STDSWHIP>1.1 or tA(runs) > 4.8)

[JMon]

Just post the f-c-ing plays!

Quit whining. Doesn't take much to copy and paste. These guys do all the work in coming up with the awesome querys.

going back through the thread, how did I miss this post. Prob a good thing I did.

<br>
<br>

[JMon]

H and -175<=line<-130 and 5.2>=tA(runs)>=4.7 and o:STDSERA<=4.2 and o:conference=AL and p:runs+po:runs<=4 and total>7 and month!=9

observe the DAY parameter

[Alex Vaile]

Looks great!

[nbilln]

4-14 (-2.67, 22.2%) avg line: 142.4 / -156.6 on / against: -$847 / +$807 ROI: -47.1% / +28.2% 7-11 (-1.17, 38.9%) avg line: -158.6 / 145.5 on / against: -$1,035 / +$888 ROI: -37.1% / +49.1% 7-11-0 (0.06, 38.9%) avg total: 8.2 over / under: -$480 / +$320 ROI: -24.7% / +15.8%

8-2 (0.70, 80.0%) avg line: -138.5 / 126.7 on / against: +$508 / -$557 ROI: +35.7% / -53.7% 6-4 (0.10, 60.0%) avg line: 123.5 / -137.1 on / against: +$325 / -$361 ROI: +27.2% / -25.1% 6-4-0 (1.70, 60.0%) avg total: 7.0 over / under: +$155 / -$255 ROI: +13.8% / -23.5%

dodgers perhaps

[BasesLoaded314]

(s:SRA + ss:SRA) = 3 and A and playoffs = 1 and season>2009 and os:SRA!=0

(s:SRA + ss:SRA)=3 and A and (month= 5 or month =7) and os:SRA<6 and oss:SRA>2 and line>-184

[JMon]

team=Cardinals and date>=20100923 and p:SRA>=6 and -8<=p:margin<=-2

[BasesLoaded314]

does anyone know how to show more than 100 entries at a time?

[cash$bro91]

does anyone know how to show more than 100 entries at a time?

This is what Mako posted in the NBA thread:

"To show more results than 50 on the page, add this to the actual URL of the browser once the query is finished loading: '+&show_games=1000'

That shows up to 1000 entries on that page for any query, and you can set the '1000' to whatever number you like."

[BasesLoaded314]

thanks cashbro!

[Alex Vaile]

Ouch Oakland, cards and dodgers all going down today with good looking querys to boot. Rough one for me. Only mets and houston pulled through. Hoping for better day tomorrow!

[BasesLoaded314]

H and month > 5 and season>2009 and 1.2 > tA(BPRA) > .6 and 8>(s:SRA-os:SRA)>-4 and (p:hits + os:walks)<15 and month!=8 and s:SRA<6 and os:SRA<5 and 30.7>(p:SIP + pp:SIP + ppp:SIP + pppp:SIP)>18.7 and (op:SIP + opp:SIP + oppp:SIP + opppp:SIP)>19.4 and total>6.4 and p: ou margin>-6.6 and streak<5 and o:streak<6 and -8 < o: ou streak<4 and os:SPT<123 and team!=Twins and team!=Red Sox and o:team!=Indians and (s:SSO + ss:SSO + sss:SSO + ssss:SSO)>9 and (s:SSO + ss:SSO + sss:SSO)<26 and (os:SSO + oss:SSO + osss:SSO +ossss:SSO +osssss:SSO)>13 and (os:SSO + oss:SSO + osss:SSO +ossss:SSO)<31

agreed. I had a rough night as well. one issue that I see with the use of SDQL to aid in in handicapping is as follows: our brains are hardwired to attempt to decode a.k.a. "see" patterns when presented with sets of data like we are given when forming queries. one must question though, whether the parameters we are using to select games are ACTUAL factors that affect the outcome of a game, or simply "white noise" that essentially has little impact on a game. ill use my query that I quoted above as a reference. This is a query that predicts unders. now I think that this is an excellent query overall. Starting on sunday, it will begin spitting out (by my estimation) approximately 100 plays until the end of the season. now in the past four years this query has been phenomenal, its been hitting at 70%......

we of course all (should) know that 70% at approx -110 odds is beyond out of this world, and if anyone had been betting exclusively these games they would be filthy rich by now. but the problem is that these games WILL NOT continue to hit at such a rate, and as alluded to before here is why: when i began to filter down this query I had same goals that are the foundation of all good SDQL queries: high ROI, high sample size and high consistency. now this query achieves a great roi and sample size, but what is sacrificed is the consistency.......

first of all, this query doesn't return results for august. why? well when I was looking at the results I noticed that it returned very mixed results in august so I simply omitted them from my results. the question is though: WHY has this situation not performed as well in august? for example, one could possibly reason that it is in the middle of the season and pitchers at this point are less focused and therefore allow more runs to score. there are many other possible explanations for what is going on here, but I would argue that its results in august are actually TO BE AS EXPECTED over a certain time frame. the key word here is VARIANCE!!!! we have all witnessed it before manifested as incredible heaters that make us feel on top of the world, as well as inexplicable cold spells that leave us feeling foolish and disgusted. that's the thing though.... over the course of time, situations arise that are out of the ordinary and results that deviate from the mean do occur....

heres another reason why this query will not perform as well in the future as before: remember what I was talking about before where our brains look for patterns in data to try to make sense and find order in a bunch of madness? well I actually (accidentally, this was a result of me going from parameter to parameter too quickly and not noticing that I had written the SDQL incorrectly) included a parameter that I am 99.99% certain does not have any bearing on the outcome of these games....

lets look at this: (p:hits + os:walks)<15
now here is what I was actually hoping to include in the query: (p:hits + os:walks ALLOWED)<15

the difference is VERY important!
os:walks ALLOWED is how many walks the pitcher in this game allowed in his previous start

os:walks is how many walks the team drew WHILE BATTING the last time this pitcher started! of course more walks correlate to more runs being scored, but these were walks issued by pitchers that ALMOST CERTAINLY wont be pitching in this game! almost always, these pitchers are from a different team than the 2 being referenced in the game
huge difference and I did this totally by accident!

but what is crazy is that (p:hits + os:walks) produces a data set that shows a very clear cut difference between games that in that past have been profitable versus those that haven't.

now (p:hits + os:walks allowed) does separate the games in a way that does show some unprofitable games that can be filtered away but the differential is much less clear. it just so happens that this COMPLETELY MEANINGLESS number JUST SO HAPPENS to have came up as a way to "filter out" unprofitable games, when in REALITY it was just a completely random aberration that I stumbled upon in the quest to boost roi....

wow this got super long, but one final point I want to mention is that in the previous years it has (in order) returned the following number of plays: 98,100,121,84,7,48,12,12,52 and 21 plays per year. no that is not a typo. in 2009, this query returned 7 results the where as in 2011 it came up with 121 results. why????? I sure as hell have no idea, I suspect it may have something to do with it being the post steroid era and all but who is to say for sure? this is just another example of how this query is a goofy as hell. how do I think it will do this year? let me just say that I will (at least in the beginning) be auto playing these results this year, and I estimate it will it will return about 10% roi, and even if it only pops out 50 plays that's still plenty ok with me. ill have turned a nice little profit and move on from there. now theres always a possibility it does very well again, and of course theres the very real possibility that this query will lose money this year. if given a long enough time frame, the ROI here WILL move down to a more realistic number, the only question is what is the "real", "stable", "average" ROI, and how long will it take to stabilize? nobody knows, if someone could figure that out they would be a very wealthy person

[JMon]

not a great night for me as well., going 3-4 -1.9

damn novel there bases, good read. I particularly enjoyed the hot/cold streak you mention of, as it happens to any handicapping method for sure. Using sdql comes all with experience (trial and error), but most importantly knowing how to profit off a particular trend, that is ride when it's hot, but also knowing to get off or dump when it's cold- ensuring a profit (much like the market). That's the hard part.

[kingvikingstad]

Hi guys, I am somewhat new to this and new to SDQL altogether. I've been teaching myself how to use it over the past week and have really enjoyed utilizing this thread. This is exactly the stuff that I love, regardless of whether I was making money on it or not (I guess I'm just a nerd at heart).

I was playing around with some queries regarding the Jays/Royals game and came across this one (albeit not applicable for tonight). The sample size is low, but it's hard to argue with 0-36-0 so I thought I'd share.

21-15 (0.17, 58.3%) avg line: 110.8 / -123.4 on / against: +$995 / -$1,235 ROI: +25.1% / -26.5% 13-12 (0.70, 52.0%) avg line: -103.6 / -109.4 on / against: +$13 / -$162 ROI: +0.4% / -5.1% 0-36-0 (-3.35, 0.0%) avg total: 9.1 over / under: -$4,005 / +$3,610 ROI: -100.0% / +92.2%

[JMon]

Hi guys, I am somewhat new to this and new to SDQL altogether. I've been teaching myself how to use it over the past week and have really enjoyed utilizing this thread. This is exactly the stuff that I love, regardless of whether I was making money on it or not (I guess I'm just a nerd at heart).

I was playing around with some queries regarding the Jays/Royals game and came across this one (albeit not applicable for tonight). The sample size is low, but it's hard to argue with 0-36-0 so I thought I'd share.

21-15 (0.17, 58.3%) avg line: 110.8 / -123.4 on / against: +$995 / -$1,235 ROI: +25.1% / -26.5% 13-12 (0.70, 52.0%) avg line: -103.6 / -109.4 on / against: +$13 / -$162 ROI: +0.4% / -5.1% 0-36-0 (-3.35, 0.0%) avg total: 9.1 over / under: -$4,005 / +$3,610 ROI: -100.0% / +92.2%

I do believe you are depicting games where it went under at home, not showing what they do after going under at home. HU is not predictive and needs a game reference.

[emceeaye]

This one looks good for today

HF and p:margin>=2 and tS(runs@team and season and site=home, N=7)>=35 and po:BL=0 and season>=2008 and o:streak<=-2 and p:W and os:streak=-2

All variables are intuitive.

[JMon]

DIV and D and p:W and 40<=o:WP<48 and team=Mets and o:team=Phillies

[JMon]

Possible under...after a game of 3 errors or more; starting a pitcher whom has a whip of 1.1 or less in his last 10.

p:errors>=3 and sA(SWHIP@starter and season, N=10)<=1.10

[Alex Vaile]

Josh Beckett goes today after posting no hitter last time out. Fade on

[JMon]

Josh Beckett goes today after posting no hitter last time out. Fade on

since 2004, this fade does have value but minimal. Easy query, can u come up with it, Alex?

[nbilln]

29-12 (1.29, 70.7%) avg line: -153.7 / 141.6 on / against: +$970 / -$1,130 ROI: +15.4% / -27.6% 16-25 (-0.21, 39.0%) avg line: 147.0 / -160.2 on / against: -$161 / -$42 ROI: -3.9% / -0.6% 23-17-1 (0.74, 57.5%) avg total: 7.9 over / under: +$475 / -$840 ROI: +10.7% / -18.4%

liking wainwright tonight

[JMon]

Good queries guys, lets see if we can put up some winners

[green7]

pitchers off very good road start

s:SRA = 0 and s:AW and HF and SG = 1 and playoffs = 0 and not C


Versus Cleveland and Kluber


The query below favors the Dodgers and Beckett.


s:SHA <4 and s:AW and H and series game and playoffs = 0 and not DIV and NGT and C and s:SRA=0 and p:HL

Good fortune tonight.

[Alex Vaile]

since 2004, this fade does have value but minimal. Easy query, can u come up with it, Alex?

I'm not sure where to start but I've been trying to. Is it SHA or s=?

And would I say team=dodgers? Or is it something like this
o:hits=0 and team=dodgers and season>=2006

[green7]

teams after losing to break long winning streak

ROI shows ad advantage to play against the team that just lost to break their streak (Blue Jays). Wanted to play ON them....won't be fading them any time soon.

p:streak > 3 and HF and not DIV and p:HL and C and conference = AL and series game = 2

[posey]

Sorry guys did not have enough free time lately.

I am backing Cueto and the Reds tonight.

STDSERA < 2 and month > 4 and o:STDSERA > 4 and s:L and ss:L and AF and league = NL
team = Reds and AF and o:team = Diamondbacks
starter = Johnny Cueto and o:team = Diamondbacks
starter = Brandon McCarthy and team = Diamondbacks

[emceeaye]

straightforward one that is good

HF and 13 >= p:margin >= 6 and s:margin > 0 and SG > 1 and os:margin<0 and tS(runs, N=9) > oS(runs, N=9) and tA(BPRA, N=10) < oA(BPRA,N=10)

Or better yet:

HF and p:margin>=7 and s:margin>0 and SG>1 and os:margin<0 and tS(runs, N=9)>oS(runs, N=9) and tA(BPRA, N=10) < oA(BPRA, N=10)

Update: sorry about that...for some reason, these queries don't paste correctly. Now they are correct.

[Alex Vaile]

HF and 13>=p:margin>=6 and s:margin>0 and SG>1 and os:margin<0 and tS(runs, N=9)>oS(runs, N=9) and tA(BPRA, N=10)<oA(BPRA,N=10)

Or better yet:

HF and p:margin>=7 and s:margin>0 and SG>1 and os:margin<0 and tS(runs, N=9)>oS(runs, N=9) and tA(BPRA, N=10)<oA(BPRA,N=10)

I tried entering the querys but couldn't get actual games just a bunch of numbers. Any trick to entering?

[JMon]

I'm not sure where to start but I've been trying to. Is it SHA or s=?

And would I say team=dodgers? Or is it something like this
o:hits=0 and team=dodgers and season>=2006

I believe this what you are looking for...
s:SIP=9 and s:SHA=0

[Alex Vaile]

I believe this what you are looking for...
s:SIP=9 and s:SHA=0

Right on JMON thanks. I was struggling with what to start with.

[JR007]

hope you guys are doing well over here

[tarosill]

HF and p:margin>=7 and s:margin>0 and SG>1 and os:margin<0 and tS(runs, N=9)>oS(runs, N=9) and tA(BPRA, N=10) < oA(BPRA, N=10)

Update: sorry about that...for some reason, these queries don't paste correctly. Now they are correct.

Nice work again emceeaye! Thanks!
I would add two additional filters here: o:site streak > -8 and WP > 48

[emceeaye]

Nice work again emceeaye! Thanks!
I would add two additional filters here: o:site streak > -8 and WP > 48

Nice addition Tarosill! even better!

[JMon]

hope you guys are doing well over here

been meaning to ask the getaway parameters of a under...and I will convert.