From a long string of Tweets from CFB Matrix
Their "Coach Effect" is a measure of how many games a CFB team wins compared to the talent of the teams playing.
Their "Coach Effect" is a measure of how many games a CFB team wins compared to the talent of the teams playing.
First Matrix rule of QB nfl draftee profiling is "are they graded"? No grade, then skip them and focus on graded guys
then sort them by the CoachEffect metric the team had while he was a starter. You want to focus on the + coacheffect guys because..
if you cant elevate your CBF team how in the hell are you doing it a NFL level? RARELY occurs.
then look at the school. Avoid guys from elite (top 10) recruiters. They bust nearly 90% of the time (vs ave of approx 67%)
The d2 guys do not have #CoachEffect metrics so watch their position in the draft. Top 2 rounds, they are usually a good bet but drafting..
a D2 or worse QB after round 3 is just setting money on fire. Good rule is 1st round of no round on d2 QBs
i'm not saying if they have high bust odds dont draft them, I'm saying draft them further down the draft because they offer less value
Before we talk about this draft lets see how these rules apply to last years draftees...
Goff: My matrix take - non elite program. 1st round draft. Massive negative coacheffect at Cal. One good yr. Round 2-3
Wentz: non elite program. D2. If you are going to draft him, he better have a round 1-2 grade. If rd 1, good odds. I'd take him over Goff
Lynch: My favorite combo. Non elite program, big coach effect #s at memphis. Elevated team. Top 3 risk to take. high grade with high CE
Cook: Graded rd 2-3. Big CE #s. Elevated team play. Top 3 risk in rounds 1-3 IMO. non elite program
Hackenberg: NFL graded round 2-4. Massive negative coacheffect at PSU. Value is in rounds 5-7. This has overhype written all over it (Rosen)
Prescott: NFL graded round 4-6. Huge CoachEffect at MSU. Non elite school. Carried team. Top 3 risk in rd 1-3.
You starting to catch on? Its quite simple. I love simplicity and strongly believe the fewer variables the stronger the correlation.