More on run/pass balance variations

[QuikSand]

I think that some of the posts here, over time, on the various "offenses" that folks have tinkered with have been pretty interesting.

We have seen that setting the run percentages pretty high (the "75 offense") can yield a very effective offense, led by a powerful ground game. We have also seen that an aggressive use of the pass on short yardage situations can yield some explosiveness to an effective offense. We've also seen some folks argue that setting an overall balanced play seection will "keep the defense guessing" and will be effective. (I'm paraphrasing from many different sources, and my well be misrepresenting some claims above-- sorry if that's so)

I'm wondering... on the matter of run/pass play selection-- do we actually gain anything out of all this on balance, or do we just shift which players get the most chances to perform?

Hypothetically, if we had an offensive team with entirely equivalent, high-quality players (however that is defined) do we believe that there is one particular set of run/pass selections that would be most effective?

Do we think that going against a normal defensive expectation (pass on short yardage, for example) is a genuine advantage? Or does it just result in a few more exciting plass plays? Do we think that a run-heavy, ball control offense is the way to go? Or does it just work best for certain types of teams?

Just curious if there is any synthesis of all the thinking that has gone on here, or if all these impressions are more or less situational. I think it's pretty interesting stuff, on balance, but I'd say that in my own career, I've felt like my successes (and failures) on offense were more easily connected to player quality than play selection. It's easy to say that, but I wonder if there's a debate to be had here...

[Morgado]

Originally posted by QuikSand:
We have seen that setting the run percentages pretty high (the "75 offense") can yield a very effective offense, led by a powerful ground game. We have also seen that an aggressive use of the pass on short yardage situations can yield some explosiveness to an effective offense. We've also seen some folks argue that setting an overall balanced play seection will "keep the defense guessing" and will be effective. (I'm paraphrasing from many different sources, and my well be misrepresenting some claims above-- sorry if that's so)

Actually Quik, now that i'm done with WLC and will only be a passive participant in GT, this was going to be the next project to tackle. Using 25 seasons or so of data on 75 Offense production courtesy of the Richmond Knights and various testing careers I have yet to run, I was going to write up something on offensive gameplanning to philosophy.

...or do we just shift which players get the most chances to perform?

Based on the experiments I did during WLC with a Midwest 35 and some other extremely strange offensive setups, I suspect this is pretty much true. A particularly interesting bit of evidence to support this is the consistently mediocre performance of outstandingly rated WRs and QBs throughout my entire WLC career. In a 75 Offense, the completion percentages and relative "safety" of the passing game (low INT, low Sacks, no screw ups, etc.) were great but the yardage and TD production was very pedestrian. Even when I had my best QBs and stables full of awesome WRs, we had almost no 1000 yard seasons and never had All-Pro QB selections.

However, the running game was a different story. The Knights led the league in rushing at least 50% of the seasons and finished top 5 in rushing for almost all of the rest of the seasons. We had All-pro RBs, FBs, and linemen on a regular basis and frequently had the league leading rusher when injuries permitted. Arguably RB Cerasani was a dominant type RB, but the rest of them were actually only Very Good running backs. Since the 75 Offense is designed to get these guys the carries and gears the offense toward running in favorable sets and situations, the RB and FB performance goes off the scale.

But take those same WRs and QBs and put them in an offense with Midwest 35 specifications and i'm sure they would light up the scoreboard and the stat tables. When I ran a Midwest 35 and a hybridized 75/35 offense, the passing numbers were good. Of course, all of this depended on having quality talent at all the right places. In the second of my Super Bowl runs, I felt I had just an adequate offensive line but dominant backs and receivers. This is what probably allowed me to go from 75 Offense to Midwest 35 at will and not lose a step.

Hypothetically, if we had an offensive team with entirely equivalent, high-quality players (however that is defined) do we believe that there is one particular set of run/pass selections that would be most effective?

I don't think so. I think what will happen is that you'll get exaggeration of stats in one direction or another. Skewing your offensive tendencies will allow you to focus on extreme quality in certain positions and slack in others. For example, a 75 Offense could theoretically get by with a dominant RB, dominant FB, three to four dominant linemen, and merely adequate players for the rest. You would need neither a stupid money QB or stupid money receivers to lead the league in rushing. So long as the QB had decent short, medium, accuracy, power, and scrambling (40s-50s), you should be okay in the short passing game that supplements the 75.

Alternately, if you go Midwest 35, you could go RB by Committee with a stable of 4 midround chumps who don't get re-signed and play RB carousel like me, Subby, and a few others. In fact, I wouldn't be surprised if you could get by with the old time RB by Committee of four minimum salary undrafted FA rookies. However, you would need a dominant stoopid money QB, about two or three stoopid money WRs, and either a dominant TE or FB with at least good to very good pass blocking linemen across the board. The emphasis in positions is different, but the result is the same - you should get offensive dominance and be able to crush the opposing defense.

I said in my WLC Round 5 and 6 thread that dominant quarterbacking is essential to a great passing team because it allows for those 15-23 yard automatic first down passes. If you play a hard core 75 Offense, you'll get maybe 4-7 yards on an average play, 9-13 yards on a good play, and maybe -1 to 3 yards on a bad play. But, if you are running a 75 Offense with a stoopid money RB who has those fabulous stats Breakaway, YPC, 3rd Down, and Short Yardage in the 80+ regions, then you are looking at consistent 8-15 yard carries on key run blocks. In a sense, you're getting short pass type yardage on runs because your back is so awesome. Given that fumbles are less frequent and usually less costly (fewer runbacks) and that ugh-ly negative yardage plays are less frequent, the 75 ground attack is almost as productive and probably safer than the aerial assault of the 35. You'll get fewer big plays, but fewer bad plays too. I think if you do an expected returns calculation on each offensive style, the risk premium benefits of running will just about offset the explosiveness benefits of passing.

Of course, I have seen a -33 yard run.

Just curious if there is any synthesis of all the thinking that has gone on here, or if all these impressions are more or less situational. I think it's pretty interesting stuff, on balance, but I'd say that in my own career, I've felt like my successes (and failures) on offense were more easily connected to player quality than play selection.

I don't think there's any argument from me here, but the thing is that your offensive system will determine *which* players will have the largest impact on the offense's success. Running a certain offense will require that your good players (since all players can't be multi-million bank breakers on a salary cap) be at certain positions. So, you could say that philosophy will dictate where you allocate your talent money. Where you have your talent will dictate what you can and can't do on offense. For example, if you had Barry Sanders and Scott Mitchell, are you going to run a 75 Offense or a Midwest 35? Since we can't have a full complement of All-Pro type players, you really have to pick and choose what kind of mismatches you want and where your offense is going to win its yardage.

I think this is why I push TE and FB so hard (even besides the fact that they're bargain positions) - these two positions are probably the only non-stupid money positions that figure prominently into all offensive styles. A running offense needs a good lead blocking FB, the TE is part of the front line blocking, and the FB and TE both provide the short pass changeup that a 75 Offense relies on. For passing offenses, the FB is often an uncovered receiver who gets ultra-high percentage passes and the TE is a good middle-deep mismatch against LBs and S in single coverage. Plus, both are part of the pass blocking complement (yep, FBs have a SkA stat too).

We've talked about whether or not dominant quarterbacking is a prerequisite to success (I don't think it is, but it makes it a hell of a lot easier heh) or whether there's any point to getting a dominant RB when RB by Committee puts up respectable numbers. I think the reason we can see situations in which one or another "dominant" position isn't necessary is because of the particular offenses and situations (often dictated by defense and special teams for field position) that frame them.

So even a team without dominant quarterbacking can win 8 of 12 Super Bowls and make the playoffs 27 out of 30 seasons... provided it has the right people for what it's trying to do.

------------------
The 64 Dollar Question: What *is* The FOF Journal?

[This message has been edited by Morgado (edited 02-08-2001).]

[Morgado]

Quik,

Hrm. Doesn't anyone ever read my line of bull? I never get any responses...

Oh yeah, just something that came to mind about the other thread and the CPU vs. human run teams issue. I think if you skip the first two or three seasons of an empty cupboard career and then look at the five years after that, it's about as close to a CPU AI team roster as a human would get. At that point, there's still holes in the depth chart but there are a couple of stars (all them top overall picks from the 0-17 seasons at the start hehe). It might be curious to see how a CPU AI run expansion team looks after say 5 years and then compare to the 5th season WLC teams.

So like if we track where the draft pick and FA signings were used and where the cap dollars ended up going, it might work like a "revealed preference" kind of thing.

------------------
The 64 Dollar Question: What *is* The FOF Journal?

[Ctown-Fan]

Originally posted by Morgado:
Hrm. Doesn't anyone ever read my line of bull? I never get any responses...

It isn't that I just find my intuitive logic cannot compete with the analysis put forth by some of the more statistically incline members of the board.

I have been hashing this thread out for a few days, but haven't reached any arguments I am happy with yet. Besides, I am sure you and Quiksand are tired of reading my stream of consiousness responses to your well though out posts.

[This message has been edited by Ctown-Fan (edited 02-09-2001).]

[This message has been edited by Ctown-Fan (edited 02-09-2001).]

[Morgado]

Originally posted by Ctown-Fan:
...the more statistically incline members of the board.

Statistics... feh. As an economist I suppose I should be defending the use of statistical methods... but oh well. Feh.

my stream of consiousness responses to your well though out posts.

Heh, you make like my posts aren't stream of unconsciousness posts... I think the reason I repeat myself so often in these strategy posts is that the little form box to enter text in only lets me see the paragraph i'm typing at the moment... so I end up repeating things I said earlier in the post because i'm too lazy to scroll up.

I really ought to write my posts offline like my dynasty reports and paste 'em in... but like I said - too lazy.

------------------
The 64 Dollar Question: What *is* The FOF Journal?

[This message has been edited by Morgado (edited 02-09-2001).]

[ez]

as usual, i offer something from the old days (i really *am* acting like an analyst emeritus, aren't i? ):
http://216.15.145.145/sideline/board...ML/000018.html

i guess it's pertinent as a look at what the AI can accomplish (relative to an active, planning human)...

[This message has been edited by ez (edited 02-09-2001).]

[Morgado]

Originally posted by ez:
...(i really *am* acting like an analyst emeritus, aren't i?)

And it's a good thing too since I (and many others I suspect) have had no exposure to the stuff on the old board. That's pretty interesting. Okay here's something I don't remember about FOF2 since I haven't played it in like 4 months - did the CPU teams adjust their training camps to eventually max out various categories of expertise?

Speaking of CPU teams, I always thought it was interesting during the WLC that every season saw the same teams in the playoffs. Denver, for one, was an absolutely dominant team and rolled into the upper rounds of the playoffs just about every season. A couple other teams like Cleveland and those damn Sturgeon kept somehow popping up near the top as well - even when their roster turnover was at a fair clip. I think your observation from the earlier experiment might hold true with getting a good young coach in FOF2001 as well (most of my WLC dominance came under one good, young coach).


------------------
The 64 Dollar Question: What *is* The FOF Journal?

[ez]

Originally posted by Morgado:
did the CPU teams adjust their training camps to eventually max out various categories of expertise?

LOL... no, they didn't:
http://216.15.145.145/sideline/board...ML/001788.html

[Ctown-Fan]

I think this is why I push TE and FB so hard (even besides the fact that they're bargain positions) - these two positions are probably the only non-stupid money positions that figure prominently into all offensive styles. A running offense needs a good lead blocking FB, the TE is part of the front line blocking, and the FB and TE both provide the short pass changeup that a 75 Offense relies on. For passing offenses, the FB is often an uncovered receiver who gets ultra-high percentage passes and the TE is a good middle-deep mismatch against LBs and S in single coverage. Plus, both are part of the pass blocking complement (yep, FBs have a SkA stat too).

I think this is an important point and needs to be emphasized. In the Midwest 35, those are two of the most important positions. You can get by with one stoopid money wide receiver and quarterback, but the offense just won't run well without a good fullback and tightend. the way I see this offense working is by spreading the field, and the te and fb will take advantage of that. How many linebackers have excellent coverage skills? In my experince, they are hard to come by, and that makes these positions critical. If you look in the NFL, teams can't really succeed without that great tightend (Wycheck, Sharpe, Coates, etc . . .). However, I don't understand why Fullbacks aren't more functional in the NFL. The Browns moved Aaron Shea to tightend, where he will excel, from the fullback position (which he played at Michigan). Now, I belive that with his skills, he would have been an even better weapon out of the backfield. I just doesn't make sense. Ok, so I am starting to ramble.