re: Doing LES with more.....

quote:

---

Again show me the data that says LSU has more "busts" than other programs.

---

for the third time...that isn;t what the OP shows.

you can keep trying to reframe the relationship in the graph, but the relationship of Les Miles getting worse offensive production with higher rated recruits versus similar schools, and in fact the average of all of FBS, is a fact.

quote:

---

All of this assumes that recruiting ranks are an accurate predictor 1:1 predictor of success. Did you account for the potential error there?

---

That potential error would apply to all teams and balance itself out, unless LSU's recuits are rated higher than they should be simply because they go to LSU. This is an argument that I've seen on the Rant about other programs, but never LSU.

Bad statistics is bad when all variables aren't the same.


Bama gets easier schedule than LSU......if you are going to do this, then you have to compare based solely on common opponents. You can't discount having to play better teams when you are looking at offensive production. Secondly, styles are a variable that completely make this point worthless based on these statistics. The only way to compare a spread offense and a run-based offense is to look at yds/play since that eliminates the fact that LSU on it's best day wants 65 snaps and a team like aTm wants 100.


If you want to compare to Bama, then use your statistical know-how to compare yds/play against common opponents and you will get your best possible comparison most likely.


Lastly, if you are going to look at recruiting rank, then you can only look at what the rank would be for each class minus defensive recruits. LSU's recruiting rank for the past couple years has been dramatically improved by defensive recruits and thus they have no direct effect on our "total offense" rank which means you are compring apples to oranges in this instance.


tl;dr - BAD STATISTICS IS BAD

quote:

---

If you want to compare to Bama, then use your statistical know-how to compare yds/play against common opponents and you will get your best possible comparison most likely.

---

Actually, the relationship being tested is offensive output to ranking of recruited talent. If you want to hypothesize that LSU produces less on offense intentionally due to style, then so be it, but the relationship is what it is.

Lulz. You really wanna say that a couple of those teams are driving the relationship in all of fbs?

Lol. Seriously, do yourself a favor and learn about leverage, then come back look at the graph again, and realize based on the spread of the points that your comment is moot.

quote:

---

Bad statistics is bad

---

Stho you are correct. While the data analysis is accurate the assumptions and judgements made by our Cpt arejust that. HIS assumptions and judgements.
Style of play , opponent, schedule etc. mostly effect offensive output.

I don't think it matters whether the recruits are under-performing or not (though I think they certainly are, due to bad coaching on the offensive side).

The offense is consistently in the bottom half of the league, it is bad by most every measure. Either Miles is recruiting good players and ruining them or he is recruiting bad players. Either way he is at fault.

quote:

---

Lulz. You really wanna say that a couple of those teams are driving the relationship in all of fbs?

---

You do realize that the spread is widely used in FBS, right? Almost the entire PAC12, Big 12, B1G, most of the mid-majors, and some of the SEC now are all spread offenses.


The point is, you are saying that Les Miles has done less with more talent.

My rebuttal is that:

1) you are taking recruiting rankings that are skewed too high for LSU because we recruit many highly rated defensive players. For you to have an accurate comparison for offensive recruiting vs offensive production, you can't use defensive recruits. Therefore you would have to recalculate rankings of recruiting classes based on just offensive players, which is damn near impossible

2) You have elected to use total offensive yards as the metric for judging us against the rest of FBS. My response was, yards don't mean anything in total when you have different styles of play. Spread offenses will always be at the top and teams like K-State, Stanford, LSU, Florida who are all successful but focus on running and T.O.P are all at a disadvantage because of style, so again the relationship will be skewed heavily toward the spread offense, hurry up teams , which covers probably 50%+ of CFB.

3) Lastly, who gives a shite about offensive yardage. If we are going to judge Miles, do it on the whole picture. We all know the offensive short-comings of the team. If we want to judge how well we've done as a program under Miles, let's look at the top 10 recruiting classes over the past 4 years and then let's look at the winning% of said teams with those recruiting classes over the past 2 years since that's when those recruits are actually on the field.

]

quote:

---

you are taking recruiting rankings that are skewed too high for LSU because we recruit many highly rated defensive players.

---

Really? Do other teams have different recruiting rules?
A good coach/recruiter gets talent for both sides of the ball

2) I will complete a multivariate in the morning using all other offensive measures. My hypothesis is you won't like those results either. LSU has neither scored nor had an exceptional TOP to significantly vary those results.

3) lmao. The miles defender argument if choice....answer me how many teams had ten win seasons this year. Then tell me how many of them are psid as much as miles.


Btw. I noticed you avoided the leverage question. Good choice as you would have gotten beat up pretty hard for not even recognizing the underlying pattern in the data.