Home Forums Skybox Sports Bar Selections Central Lagniappe's Stat Busters

This work is based on that of Todd Beck, who owns the copyrighted site thepredictiontracker It is a great site and I really appreciate Todd's efforts in maintaining those tables.

My added value is that I take the top linesmen (must be over 53% ATS) and average them. I group-source among the best. A similar method has proved to be highly reliable in projecting baseball players performances, I have used this technique for three years, and it has been useful enough to continue it.

I like to find college games where the consensus is a little better than the line. Conventional wisdom is the line is tilted towards the "public" teams. Then I take this list and find the games where the "best" line is better than the overall consensus line. This, of course, assumes the winners will keep winning. The top 9 changes frequently throughout the season, so "it ain't necessarily so".

I look closely at games where the consensus and best deviate. There may be hidden value in the line in those games (injury, weather, trend, etc).

I will also submit the NFL work, but the Vegas line may be sharper overall then the real line. I do use these results in ATS confidence pools, however.

Notes for Week 3 College, Week2 NFL
If you download the file you will see the games by home-team alphabetical order. The prediction averages, means, and win probabilities are directly from thepredictiontracker.com site. The Best Columns are from the lines of the most accurate ATS of the total in the group. The best linesmen and then listed with their line for each game.

Sheet Observations:
Best likes Fla St, Memphis, Utah, Florida, Minnesota, NC State, Virginia Tech, TCU, Washington better than the consensus and better than the line.

Consensus likes Colorado State, North Carolina and a lot better than line and a little better than best

ND-GT is schizoid among linesman. Uncertainty on how to rate Kizer. Toledo-ISU, UCLA-BYU, WSU-Wyo also foggy.

Best likes Oregon and Wisky almost a TD more than consensus better thanthe current line, but I don't lay 35 points.

Prior sheet didn't open, this won't be a problem going forward.

Attached is what the spreadsheet had in it. Click to view.

Sorry for the issue, it's my fault.

Last weeks results: The "best" line last week led to a 6-1-1 record. Winners were FSU, Utah, Fla, NC State, VaT, UW. Loser was Min and Push was Memphis.
Consensus unconfirmed by Best went 2-0 (NC and CSU). TCU, Wisc and Ore had too many points for my involvement.

The nine lines representing the "best" has two changes. Dropping out is Born and ThompsonCAL. Joining the list is Thepowerrank.com and The Sports Cruncher.

This week's observations: Alabama, Baylor and LSU overpriced (but if Dungey doesn't play, LSU is cheap). Can big dogs continue to rule like Week 3?

Best and Consensus think the line move in Ark v. TAMU is in wrong direction. Bet into it (take Ark).
Conflict between Median and Average in BC v. No.Ill is warning that different linesmakers handled BC QB situation differently. Avoid unless you love little Flutie or have more insight. Second of three road games for Huskies.

Best and Consensus like Charlotte, Marshall, Missouri, BYU, CFU, Washington and (yuck) Fresno.
Boise also qualifies, but for injury and travel reasons, I arbitrarily pulled them from the qualified list.

As always thanks to Todd Beck, and his great site thepredictiontracker I really appreciate Todd's efforts in maintaining those tables.

Additional Notes: Charlotte is not a play if QB Johnson plays for the Owls.na, Kansas, and So. Alabama. Check injuries on each of these games before playing. If skill player is out, scratch from list.

To see the information you must click on the attachment. I think you'll like what you see.

Last week was a reminder that no system has yet been conceived that can always beat a randomly generated string of variables. We were routed! The system is as follows: Play on any team that has a three point edge with the "best" and a number between the line and the Best by the consensus. This applies only to games where the spread is 21 points or less. Week one results were: The "best" line last week led to a 6-1-1 record. This week, the system produced a 2-6 record. With injury and travel filters I pulled out two losers and one winner (Boise), because of injuries, so our record is a paltry 1-4. Season record is now 7-5-1.

I am changing the presentation a little this week. If a team has unknown injuries on Thursdays when I post I will scratch out their line and that game is "off of our board". Also if the variance (sigma) exceeds the spread, the game is not usable, because the difference in the "hot lines" is so extreme.

As is the case in a breakdown the nine best lines services are jumbled. Without comment on the games, the system plays this week in college are Idaho (Ark St), San Jose (Auburn), Tulsa (Houston), and PItt (VTech). These are all dawgs and all ugly, but that is what the spreadsheet gave this week. The theme is that yucky teams are discounted too deeply.

A note on the NFL table. The best handicappers like Pittsburg, even without Big Ben to win the Thurs nite game

Here is the spreadsheet.

I went back to week 2 as a starting point and looked at games that showed a chance of covering the spread 56% or higher.

Week 2 went 2-2
Week 3 went 3-2-1
Week 4 went 3-2-1

Overall 8-6-2 for 57.14%

When looking at the record for over 60% the record was 2-2-2

With that said if you look at 56% to 59.99% the overall record is

6-4 for 60%

8 games qualify under my system this week.

Buffalo 60.2
Illinois 58.1
Indiana 56.8
Michigan St 58.1
Northwestern 59.7
TCU 56.7
UCLA 56
UTEP 57

Keep in mind this stat breakdown is on the home team only and doesn't include away teams. There may be a team or two along the way considered the home team on a neutral field.

I didn't break it down by dogs/favorites but may look into that.

At this point anything hitting 57% + is worth a look no matter the parameters. We will have a bigger sample size after this week.

To add to the above

Favorites 6-5-2
Dogs 2-1

5-3 for the day.

Overall 13-9-2 -- 59.1%

Over 60% -- 3-2-2 overall
56-59.9% 10-7

Favorites 2-2
Overall 8-7-2

Dogs 3-1
Overall 5-2

The spreadsheet is pointing to a lot of games this week. As always, check injuries and weather before buying any tickets. The arithmetic is oblivious to those factors. This weeks qualifiers: Home Faves--s, Marshall, Toledo; Home Dawgs--UNLV (CHECK INJURIES!), Kansas. Away Faves-- Syracuse; Away Dawgs--UConn, UTEP, Wisky, Ball St, Navy, Indiana, and CMU.

We can add Washington, Maryland, Pitt, and ECU to our watchlist. Lines move and more games become playable. The key to a game moving into a criterion for a play is whether it moved because of injury. If it did, ignore it. If not, the odds favor betting into the line move. If we can't get the Updated number in the chart (or better), then it falls off the list.

Line plays went 13-6 this week. Home favorites went 3-1, Home Faves 2-0, Away Dogs 9-5. From the chart there were 5 Injury no plays.

Winners were Washington, Marshall, Maryland, Toledo, CMU, Wake, Troy, WSU, Pitt, Tulsa, K-State, Wisky, ULL and SDSU. Losers were Indiana, UMass, Navy, UConn, Syracuse.

Injury No Plays wer VatEch, Ball St, NMSU, Kansas, and UNLV

SEASON RECORDS

Last week 13-6.
Overall 26-15-2 -- 63.4%


Plays this week are Ball state, Marshall, UNLV (if Decker OK), Memphis, Michigan, Ohio, UTSA (if Bogenschultz passes concussion protocol), Syracuse, Washington, and Purdue.

REVISED PLAYS FOR SPREADSHEET ABOVE:
Missouri, UTSA (if Bogenschultz passes concussion protocol), Purdue, Syracus, Marshall, Memphis, Ball St, Michigan, Washington, and OHio.

NFL plays would be Atl and Den. Will not include NFL in our W-L compilation records

Last weeks plays went 5-4. Faves were 1-3 and Dawgs were 4-1. Winners were Marshall, Memphis, Syracuse, Purdue, Missouri. Losers were Ball, Michigan, Ohio, and Washington. Injury no-plays were UNLV and UTSA.

That brings season record to 31-19-2 (62%)

This weeks plays are Illinois, Toledo, Army, Rutgers, South Florida, and Memphis.

I don't know how I missed it, but Temple is also a play. So the fav pics are Temple, Toledo, Memphis and the dawgs are Illinois, Army, USF, and Rutgers.

Last week was a sweet 5-2. Winners were Toledo, Army, So.Fla, Memphis,Temple. Losers were Illinois and Rutgers. Faves went 2-0 and DAwgs 2-2.

For the season that makes us 36-21 (63%).

This week we only have two plays--Iowa and Stanford. Marshall also qualifies on the chart, but leading RB Johnson is out, so the lines have correctly adjusted for that. Good luck this week.

Hawkeyes have just announced that RB Jordan Canzeri will not play. I am still good with the play because Weisman, Wadley can carry the load.