SDQL Trends for Publication
Joe Meyer
Our friends at KillerSports.com publish an NFL Annual each year.
This year they are soliciting SDQL trends for inclusion in their Annual.
If your trend is selected you will be offered your choice of a free
1/4 page ad in the annual
or $100 in web debit value at KillerSports.com and KillerCappers.com
The top 10 SDQL trends will be selected.
To enter your simply send your SDQL text along with the English
translation to this group.
Best wishes,
Joe Meyer
Farbror the Guru
Nice! I will start off with a simple one: "The Trend is your Friend" -- Bet OVER if the home team is dog AND the home team went over last week
p:O and season = 2004,2005,2006,2007 and D
This one is also pretty nice: "Defence in week 1" -- Bet UNDER in week 1
week = 1 and season > 2003
UNDER is also a pretty good bet in week 2
There is some logic behind both trends:
1. The BettingJoe's are zig-zagging too much
2. All teams are really focused early in the season and try to focus on defence and simple plays
Livsmannen & Donatorn
www.farbrortheguru.com
> Date: Wed, 15 Jul 2009 21:30:50 -0700
> Subject: SDQL Trends for Publication
> From: jam...@sportsdatabase.com
> To: SportsD...@googlegroups.com
Chris Platt
The first query is excluding 2008, which went 14-18. The query is also suppose to include the home team, which is missing from the SDQL. If you look at all seasons, this query (HD and p:O) is only 52%.
I messed around with the second query and came up with this:
week=2 and p:O and total<40
Bet the under on teams in week 2 when the line is less than 40, off a game in which they went over. This query is 69-34 (67%) since 1989. The problem is this query includes games listed twice, anyone know how to simply remove the duplicated games? I manually removed the duplicates and the trend still holds at 65% (58-31).
> </html
Joe Meyer
I couldn't reproduce the record on your first trend.
Can you please send the full SDQL and English translation?
I like your Over Under trend and you need to get rid of the duplicate games.
For Over/Under trends it is often useful to look at only half of each
game by setting:
site = home
(or the short cut `H`)
Then you have to check for the away team by hand using the 'op'
prefix like this:
week=2 and (p:O or op:O) and total<40 and H
Please verify this query by checking the game listings and insuring
that this query is giving exactly what you want.
Then resubmit the SDQL along with your English translation and I'm
sure this trend
will be ranked high.
The top ten ranked trends as of August 1 will be given their choice of
a 1/4 page ad
in the widely distributed, long publishing, and best-of-class NFL
Annual or $100 in web debit value for http://KillerSports.com trend
libraries and custom query methods and/or http://KillerCappers.com
SDQL trends and selections.
Best SDQL wishes,
Joe Meyer
Farbror the Guru
I read somewhere about the following system: Bet over the highest total on the board and bet under the lowest total
Rational: The bookie cannot post a proper nuber on the extreme games becuase of the risk of a totally loopside betting action
Is there any way to check this idea using the SQL code?
Cheers, Patrik
Livsmannen & Donatorn
Patrik Öhagen 018 - 244 116(h)
Hamnesplanaden 4D 018 - 174 924(a)
753 19 Uppsala 073 - 92629 53
Skype: farbrortheguru
Gör personlighetstestet på MSN Dejting, se vem du passar ihop med! MSN Dejting
Chris Platt
The first trend wasn't a submission or anything. I was just commenting on Farbror the Guru's SQDL. The SDQL that I referred to as winning 52% was simply "HD and p:O", which is 353-321 on the over. The 14-18 record was the 2008 season result, which was excluded from Farbror the Guru's SQDL.
Thanks for the help on other query, that is exactly what I needed! So the SDQL submission is "week=2 and (p:O or op:O) and total<40 and H" and the English translation is "home teams in week 2 when the total line is less than 40 and either team went over in their previous game", which the trend would be to bet the under in this situation as it has gone 58-31 since 1989.
Thanks,
Chris Platt
Joe Meyer
the SDQL for the `total was the highest of the day` is:
Maximim(total@date+1)[date]=total
Note how the summative is pushed ahead in time when calculated with:
@date+1
so that it is ready when evaluated at the current date using the
square bracket notation:
[date]
Matthias Wiermann
This is very interesting. I have run this for the moneyline:
Maximim(line@date+1)[date]=line
Worth noting that this shows a positive return at 142-238 +$1,105
without adding any further conditions. However I don't have an
explanation why for some dates there are no results - e.g. between June
25 and 30. My initial thought was that it shows only the maximum line of
the day when there is just one clear max line. Yet on June 24 there are
even four games listed. Any explanation for this?
Cheers,
Matthias
Joe Meyer schrieb:
Farbror the Guru
Livsmannen & Donatorn
Patrik Öhagen 018 - 244 116(h)
Hamnesplanaden 4D 018 - 174 924(a)
753 19 Uppsala 073 - 92629 53
Skype: farbrortheguru
> Subject: Re: Possible to check this idea??
> From: jam...@sportsdatabase.com
> To: SportsD...@googlegroups.com
Joe Meyer
To find the games where the line was the highest of the day use the SDQL:
Max(n:line@n:date)[date] = line
Farbror the Guru
Joe, Thanx! This one seems to work for MLB but I cannot make it work for the NFL. What did I miss?
Livsmannen & Donatorn
Patrik Öhagen 018 - 244 116(h)
Hamnesplanaden 4D 018 - 174 924(a)
753 19 Uppsala 073 - 92629 53
Skype: farbrortheguru
Joe Meyer
you may need to spell out 'Maximum'
Best wishes,
Joe Meyer