SDQL Query of the Day

SDQL Query of the Day Gentlemen, I'm going to try to post an NFL Query of the Day in this thread. The SDQL text box is available from a link on the lvasports.com home page, so to run the query yourself, simply cut the SDQL text below and paste it into the SDQL text box. [B]p:penalties=2000 [/B] The p: prefix points the parameter immediately following it to the previous game. So the English text for the above SDQL text is: The Cowboys since the 2000 season when they are off a game in which they committed five or fewer penalties and did not commit a turnover. To see the complete game listing, click on the SDQL text to the far right. Note that the purpose of posting these queries is NOT to suggest plays. It is to introduce those who are interested to the Sports Data Query Language -- the most powerful tool on the internet to investigate past results in the NFL, MLB and the NBA. If you have a query that you would like performed, post it in a separate thread and I'll respond. Prof Meyer[QUOTE][QUOTE][/QUOTE][/QUOTE]

8.23.10 The Cardinals since 2002 at home when they lost by less than a TD in their last game. The SDQL is: [B]team=Cardinals and site=home and -7=2002[/B]

Wow, very powerful tool for datamining!!! I might need a bit of help regarding these queries: 1) How do you seek out intra-conference games (not same conference) eg: home favourites of 4 or more, who lost last game and is playing an opponent from the other conference 2) Is it possible to seek for starting time? I would like to examine west goes east early playing time eg: OAK/SD/SEA/SF as a dog or favorite of not more than 3 points, playing a 1 oclock starting time on the east coast 3) Are searches made on opening or closing lines?

[QUOTE=Superskrub;25565] 1) How do you seek out intra-conference games (not same conference) eg: home favourites of 4 or more, who lost last game and is playing an opponent from the other conference[/QUOTE] [B]H and line

Lines? [QUOTE=Superskrub;25565] 3) Are searches made on opening or closing lines?[/QUOTE] The lines are entered by our man in Vegas. They are the lines at which most bettors made the play. So, they are neither openers nor closers.

[QUOTE=Superskrub;25565]Wow, very powerful tool for datamining!!! 2) Is it possible to seek for starting time? I would like to examine west goes east early playing time eg: OAK/SD/SEA/SF as a dog or favorite of not more than 3 points, playing a 1 oclock starting time on the east coast [/QUOTE] At this time, we can't differentiate between a 1:00 start and a 4:00 start on Sunday. We can examine everything else. So, the SDQL below: -3

8.24.10 SDQL Query of the Day 8/24/10: The Ravens in franchise history when they blew a halftime lead and lost at home last week. SDQL Text: [B]team=Ravens and p:HL and NB and p:M2>0[/B] Note that NB is a shortcut for 'Not off a Bye" When you run this query you will see that the Ravens are perfect in franchise history in this spot, covering by an average of over two touchdowns per game. To see the list of games, click on the link that appears to the right after the query is performed. Prof Meyer

Do you bet any of these 9-0, 6-0, 11-0 teambased stats? Because when you take away the team from those queries, you will often get something that goes very close to 50% with 3-400 games. I have started listening to this "Behind the bets" podcast. I think its 3-4 episodes back that this statistics guy said that he never bets teamspecific statistics, as you will never get to at least 100 games in a 3-4 year period that he feels comfortable with being statistically sound.

Whats dpa and dps short for?

[QUOTE=Superskrub;25618]Do you bet any of these 9-0, 6-0, 11-0 teambased stats? Because when you take away the team from those queries, you will often get something that goes very close to 50% with 3-400 games. I have started listening to this "Behind the bets" podcast. I think its 3-4 episodes back that this statistics guy said that he never bets teamspecific statistics, as you will never get to at least 100 games in a 3-4 year period that he feels comfortable with being statistically sound.[/QUOTE] When handicapping, I look at a LOT of trends in past performance. I have no delusions whatsoever that an 8-0 trend has something close to 100% of continuing. When I write up a game with play-on trends of 13-0, 9-0 and 9-0 for one team and play-against trends of 0-14, and 0-11 for their opponent, I'm hoping that I have, say, a 57% chance of winning. When evaluating the "strength" of a team trend, I ask myself, "Does this team trend fit into the character of the team?" I ask, What are the margins by which the team is covering or not covering?" I ask, "how timely is the trend?" I ask, "Is there an assignable cause for this trend?" I have no problem playing against a 14-0 trend if it not timely, it doesn't make handicapping sense and/or the margins are small. When a team covers seven straight in a particular situation by at least a TD, that gets my attention. All other things being equal, I would prefer a 7-0 trend with a huge margin rather than a 14-0 trend with a few close covers. Finally, I disagree with the "statistician" who needs 100 games before he plays a team trend. Of course, this would be ideal. But handicapping football is not like rolling die or flipping coins. It would be nice to run the experiment 500 times, but we just can't do that. We have to work with the data we have. We have to draw the best conclusions we can. When I quote a 23-0 ATS trend in a write-up, I'm saying that because this team has covered 23 straight times in this situation, we can have a positive expectation value if we play on it to continue. That is, the team is likely to cover again. I just don't understand how some handicappers can purposely ignore a team's recent results simply because they did not play enough games. I have no problem using a trend that is "only" 3-0 ATS if it makes sense, all three games are from the current season and all three solid were double-digit covers. It's not a perfect handicapping system. I'm not going to pick 100% winners. Yes, the trend might be a random fluctuation. Yes, we might lose. But it's the side on which to be. And that's the goal. Get on the right side. To answer your original question, I would not base a play on a single 8-0 trend by itself simply because I have so much more information at my disposal. A recommended wager is always based on a thorough investigation of the trends, systems, injury reports, newspaper reports and more. THEN I ask the most important question of all -- AM I GETTING A GOOD ENOUGH LINE? Only if all the criteria are met do I release a play. The trends are pieces to the puzzle -- not the entire picture. Prof M.