Another take on the dumpoff debate

[Jacko1971]

For those of you still with me that haven't switched off at the word "dumpoff" here is a monster posting on the topic!

If you are anything like me in the way you approach GP, and DP in particular, then...

- when DP is completed against you, you become incandescent with rage!
- but, if your team completes DP, you conveniently forget how annoying it is to be on the receiving end (no pun intended)

- if you lose a game to a team that has completed a lot of DP, you blame the GP engine for the loss,
- but if you win a game with DP you praise your secondary receivers for staying in their patterns and getting open!

So, in an attempt to address this topsy-turvy balance and arrive at some sort of reasonable, final conclusion (so that I can put DP to bed and never darken this forum again with my ramblings about it) I've been looking at some numbers.

First of all I looked at NFL stats for this past season to see how many 20+ yd passes QBs throw per season. I don't necessarily think GP has to match the NFL, but if the NFL stats are anywhere near those that GP generates then we can't moan about GP, and that's the key - if the NFL shows "dumpoff" type stats similar to GP then all is hunky-dory. The top 24 QBs in the NFL had an average of 47 completions of 20 or more yards. So how does this compare to Gameplan? To make a comparison all I can do, of course, is look at what happens in my own games, so I took my three advanced teams and looked at the seasons recently completed. I looked at the amount of "normal" OPL completions and WR dumpoff completed against my defence. (NB, these are my only seasons playing Gameplan, and if I am particularly bad, or good, against OPL and dumpoffs then the data will obviously be skewed, so I would be interested if anyone could compile similar stats over many seasons).


NFLAR Bears (2-14)
NFLAY Rams (11-5)
NFLC Bills (9-6-1)

So that gives 48 games of data for teams with a combined 22-25-1 record - a nice "average" sort of number.

And this is what we get, for those 48 games averaged out to give a 16 game season. This data is taken from stats breakdown section of the game report.

BASE OPL STATISTICS - FACED BY THREE GP ADVANCED TEAM DEFENCES
OPL ATTEMPTS: 161
OPL COMPLETIONS: 35
PERCENTAGE: 22%
YARDS: 869
COMPLETION AVG: 25.1

WR DUMPOFF STATISTICS
COMPLETIONS: 46
YARDS: 835
AVERAGE: 18.2

NB: I have purposely omitted RB and TE dumpoffs.

The dumpoff data above is DP and SP, as it is not possible to split between the two. SP obviously reduces the average, but since the 18.2 average is towards 20 we can reasonably assume that quite a high number of those 46 dumpoffs are over 20 yards, and we know from experience that DP regularly gets over 20 yards. Also, not all of the 35 "normal" DP completions will be over 20 yards, despite the 25.1 average. DI, DO bring the average down, DL increases it. So it is proving tricky to generate a number for 20+ yd passes in GP.

But let's try any way! If we were to suggest that about two-thirds of normal OPL and dumpoff completions are for 20+ yards then my data results in:

OPL 20+ = 35x2/3= 23
dump 20+ = 46x2/3 = 31
total 20+ = 54

Hence a total of 54 passes over 20 yards for an "average" 8-8 type team, which isn't a million miles from this seasons NFL average of 47. If my two-thirds assumption is too low, then the number rises, if my assumption is too high, the number reduces.

What this all means to me is that I'm not quite so angry about dumpoffs as perhaps I was before, so if the stats breakdown shows I'm up against a team that averages up to half a dozen big completions a game, then fair enough. What it also means is that I am not so averse myself to employing a bit of S and D in my own offences now! I suspect however that the "spread" of values around the league is probably quite wide - significantly wider than in the NFL. GP run and shoot teams will probably have a lot more 20+ completions than the average, and running teams a lot less, so it will still be annoying to lose to a team that has double-digit DP completions in a game.

Finally, I appreciate this is all quite anal! But, if anyone else has their own numbers that they'd like to share to see if they are similar or otherwise to those I have compiled it would make interesting reading (to me any way!)

[Darren Halford]

I have never known what all the fuss is about concerning dump offs!
It happens in the NFL,so why all the complaints?
I'm sure,although i don't have the data to prove it,that it all evens itself out over the course of a gameplan season,you win a couple of games with DP help and you may lose a couple of games due to DP.
Get over it!
Very intresting read all the same.

[Jacko1971]

you win a couple of games with DP help and you may lose a couple of games due to DP.

Ah, but you don't - it doesn't even out, at least not for everyone, because in order to really exploit DP you need to be running formations and plays that lend themselves to it, hence this is where the perceived unfair advantage occurs. Of course we could all just go S and D with DS, SL etc and then it would even out, but if stretch offences are not your bag then it feels alien to use them.

[Alex Coombe]

I think the real bugbear with DP is that gameplan is a matter of matching a good defence to the right offensive play and vis versa, whereas DP is the only part of the game where your opponent's play can go the distance against your brilliant defensive call despite you 'getting it right'.
You don't see PW RD breakout for a 70 yard TD in GP, but you do in RL, great players make plays, sometimes safeties miss a tackle etc
You don't see RO OV with the HB 'escaping' for a 70 yard TD the same way a S QL ND, S SI LD, D SO ZS can DP for a 70 yard TD.

Both my teams have ORI ORO OPS OQB DPR rbs and I have been extremely dissapointed in the ability of the running game of either team to get big yardage against anything other than BZ WS FS FD IX OX, the running game doesn't seem realistic to me, it is rock paper scissors, but the passing game isn't because you can put your rock against their scissors and the pass is completed to paper!

So although the number of DPs and deep completions may be representative of RL I think it will continue to annoy people because of the fact that it is not a primary call, therefore it is less like real gameplan, where you match up the right plays.
I'm sure peoiple will reply saying you can limit it with certain play calls, which is true, but compared to stopping a RT or SW it is ridiculously hard IMHO.

So as a result people who build high octane passing offences gain the benefit of having this backup ability to succeed even when they get the call wrong, whereas everyone else gets a bad matchup and gets stuffed, sacked picked off etc. I think it makes 4 WR formations too strong, which makes many people resort to using them, which only serves to infuriate the people who prefer a different style, or those who don't wish to frustrate their opponents by exploiting this ability to get a good result from a bad call.

I think a better option would be to drastically reduce the number of DPs and increase the number of primary completions to keep the overall numbers realistic, especially against poor matchups,like RD and GL for example.

[Andrew Woodlock]

Good post Alex

Both my teams have ORI ORO OPS OQB DPR rb's and I have been extremely disappointed in the ability of the running game of either team to get big yardage against anything other than BZ WS FS FD IX OX, the running game doesn't seem realistic to me, it is rock paper scissors, but the passing game isn't because you can put your rock against their scissors and the pass is completed to paper!

I have just got my HB up to ORI ORO OPS OQB DPR and am quite excited about the prospects but as you rightly point out he is unlikely to produce anything special against a coach who runs 30 RD defensive plays. Where as in another league my 5* QB can complete DP's and primary passes against ZD, WC, ZS, ND, PD and any other Defence you wish to name.

That's why I have started introducing RO into my Gameplan as it gets decent if unspectacular yards against RD.

I could run a RO RD with a 1* HB and get 4/5 yards where as I could run a PW RD with a 4*(DPR) HB and get 2 yards. This I could accept if occasionally he broke one between even 15-30 yards now and again wth the talent he has!

I know people will criticise but I would not want to base my Gameplan around running 45 times to get the Defence RD etc and then throw DL/DS which inevitably gets dumped off as a DP!

JMO,

Andy

[Dan Slavin]

or those who don't wish to further sully the game by exploiting this ability to get a good result from a bad call.

I think a better option would be to drastically reduce the number of DPs and increase the number of primary completions to keep the overall numbers realistic, especially against poor matchups,like RD and GL for example.

I'm on board with a lot of what Alex is saying. I don't think increasing primary completions will work. Presumably in these instances the D will get the right matchup and it still completes. That'll just annoy people more than now. I'd increase very short dumpoffs to the TE or RB.

I had a WTF moment when I read the 'sully the game' comment! I just thought "What a sanctimonious, 'holier than thou' comment is that? Who does this guy think he is? Who made him the moral compass of GP? Why does he think it's acceptable for him to go around accusing people of sullying the game?" What a shame that an otherwise very good post had this type of comment tacked on the end.

[Thomas Hennes]

Great post, Alex !

I have nowhere near the GP experience that most of you have, but having played about 20 years ago, and then taking up the game again very recently, it was quite obvious to me that the biggest change - by far - was the introduction of dump-off passes in the form of actual plays (I believe they existed before, but were not actual plays and could not be trained - I vividly remember some plays being dumped to a HB on the flat, for instance, maybe someone can confirm this). DP attracts the most scrutiny because, like Alex pointed out, it has game-breaking ability and gains yardage in massive chunks.

As a matter of fact, the S/D debate can be reduced to the DP debate, as the stretch wouldn't function quite as well without the big-play threat of the DP, and, more importantly, its ability to turn a losing match-up (pass defence vs. passing play) into a winning one ("checked-off to WR for 37 yards").

Where Jacko makes a valid point nonetheless is by pointing out that the DP is not unrealistic in statistical terms. My belief is that the DP (and other dumpoffs) was introduced as a mean to improve the passing game - I remember the dominance of run-oriented offenses (and big defenses) in early GP - and make pass-oriented offenses (the blueprint for success of the "real" NFL nowadays) a viable option for GP coaches. The only problem is that it apparently didn't fix the balance (an all-important concept in any game!) between run, pass and defense, but only shifted it in favor of passing & blitzing where it used to be in favor of running & run defense.

What makes GP a fantastic game is the play-calling match-ups between two live coaches, each trying to outsmart and outfox the other guy, and putting in some thought and analysis trying to detect trends etc. in the opposition's style of play. There is great satisfaction in matching your opponent defensively and finding mismatches to exploit offensively. What makes DP such a bad play, in my opinion, is that it overrides that very satisfaction. It is very deflating to call the right defense just to find yourself beaten by an unlikely (and unrealistic, in most cases) deep dump-off ; and it is not quite as exhilarating to edge your opponent thanks to a few critical DPs in situations where you were clearly beaten and the other guy had the right defensive play on.

I like Alex's suggestions as well ; make medium-to-long passes a higher % rate against obviously committed run defenses, and quick & short passes a higher % rate against deep pass defense or blitzes. Also improve the running game slightly by making breakouts more likely against "stacked" run defenses (have 8 or 9 players in the "box" is very risky in real-life football - you can stuff or stop the runner in most cases, but if he gets past the first line he's likely to go all the way). As it is, defenses like GL should never be called in standard situations (ie anything thats not either goal line stands or 4th & inches situations) because of the high breakout potential that SHOULD be involved.

[paperbackwriter]

I've always enjoyed Alex's posts, but Thomas is coming up fast on the outside as New Favourite Poster. Great stuff from both, and I agree with it all. And go easy on Alex, Dan - "sully" might be a poor choice of words but I don't think he was setting himself as the moral arbiter!

[Alex Coombe]

all just my opinion, lets pretend I said detract from, rather than sully.

There is great satisfaction in matching your opponent defensively and finding mismatches to exploit offensively. What makes DP such a bad play, in my opinion, is that it overrides that very satisfaction.

I definatley agree, and I think the more DP there is in a game the more it detracts from the this enjoyable coach vs coach matchup.

Therefore I find the more times I play an S or D heavy offence the less I enjoy gameplan, because the result of the game is linked less to the interaction of my gameplan with my opponents.

[paperbackwriter]

I'm with you on that one, Alex.

[Jason O Mahony]

all just my opinion, lets pretend I said detract from, rather than sully.

I definatley agree, and I think the more DP there is in a game the more it detracts from the this enjoyable coach vs coach matchup.

Therefore I find the more times I play an S or D heavy offence the less I enjoy gameplan, because the result of the game is linked less to the interaction of my gameplan with my opponents.

Interesting Alex that you constantly use the S or D when referring to DP. I have played a few teams/coaches who never called S or D formations once in there Offence Gameplan but yet end up with as much of, if not more, DP yardage than what my S/D Offence had. Even this past season in AY when we both faced each other during the regular season, I seem to recall you had an issue with the amount of DP my team had.....but if you look a little closer at the stats....I think your DP was less than 20 yards or so behind ??!!??

My point basically is....you don't have to box away the DP issue with JUST S & D formations.......you could throw in Z & C & B and possibly a few more.

All just IMO.

[Alex Coombe]

I find it much worse from the 4 WR sets than the 3s, IMX it is rare from 2WR sets and not overpowering in the Z C H.
Back on the OP's original topic I find the level of DP from say the A formation to feel realistic.

We might have had similar amounts of DP in that game, the difference perhaps is I was trying all out to do everything I could to limit yours, wheras I doubt you could say the same of my offence.

[Jacko1971]

This is superfluous, because GP won't be changed, but a change I would like to see is that passing from 4 WR formations results in more hurries, that way there is a better pay off for D. This, with short dumpoffs, and the ability for star backs to break decent gainers even against run defences would make GP so much better than the good game it already is.

On the RB question, I often see the words "breakout" against, say, a four yard gain! Not really a breakout eh? Keep the percentage of breakouts, but treble the yardage perhaps?

[Alex Coombe]

i believe the rulebook says that Breakout means someone on D missed a tackle.

[Alan Milnes]

I think the problem with dump-offs is simply the terminology - rename it "completion to secondary receiver" and people would understand what was going on and get much less wound up about it. If you call RT you are reliant on your running back making the play - if you call S DL you have 4/5 receivers running routes and you only need one to get open.

[Andrew Woodlock]

if you call S DL you have 4/5 receivers running routes and you only need one to get open.

Correct, though my original annoyance was that the DP killed me against pass type Defences from the 'I' formation?

[Alex Coombe]

If you call RT you are reliant on your running back making the play - if you call S DL you have 4/5 receivers running routes and you only need one to get open.

There are lots of other guys who can make a play, a textbook WR block for example is the reason for many many big breakout runs on an innocent looking sweep...

[Alan Milnes]

If you call RT you are reliant on your running back making the play - if you call S DL you have 4/5 receivers running routes and you only need one to get open.

There are lots of other guys who can make a play, a textbook WR block for example is the reason for many many big breakout runs on an innocent looking sweep...

Well yeah, but the same can happen on a pass play as someone blocks you open. I guess it's perhaps a philosophical difference here.

[Alex Coombe]

jacko has done the groundwork and found the length of pass plays is realistic, perhaps we need data on RB runs (and gadgets too!)

[Alan Milnes]

Good idea - gut instinct: gadgets are too successful.

[paperbackwriter]

That's definitely true, Alan.

If EV and RV were weakened, it would remove a staple from the S/D teams who run SL DR DL EV or variations, and would make it much easier to defend. I'd be all for it.

[Alan Milnes]

Yes - thinking about that is probably a lot of the frustation of playing against S/D - you have to respect the run far more than you should.

[Mike Thorp-Potter]

I actually think that EV is way too succesful in Gameplan, and not realistic to real life results.

You don't see NFL teams running EV 5-10 times a game and getting 50 to 70yd TD's very often,
but it happens in Gameplan.

Alan has a very good point about the effect that EV has on respecting the run from S & D formations,
it is not realistic in my opinion, and is much more of a concern with S & D formations than the dumpoffs.

[Andreas Horch]

I actually think that EV is way too succesful in Gameplan, and not realistic to real life results.

You don't see NFL teams running EV 5-10 times a game and getting 50 to 70yd TD's very often,
but it happens in Gameplan.

Alan has a very good point about the effect that EV has on respecting the run from S & D formations,
it is not realistic in my opinion, and is much more of a concern with S & D formations than the dumpoffs.

I would say so too. What I don't understand is: The rulebook says reverses are "very risky against blitzes", but according to my observations, they seem to work 9 out of 10 times vs my blitzes and generate lots of yards.

[Alex Coombe]

rulebook also says the trickier plays wear out quickly as the defence starts to expect them, while basic plays like RT don't wear out much.
I've rarely noticed EV wearing out.

[Jon Gick]

If EV and RV were weakened, it would remove a staple from the S/D teams who run SL DR DL EV or variations, and would make it much easier to defend. I'd be all for it.

I totally agree that the reverses are too powerful in general and could do with being tweaked. However if they were weakened and everything else left as it is, I'd suggest that it would make all single back formations (apart from Z*) too easy to defend. Basically, reverses seem to me to be the only reliable way to beat the BZ when you haven't got both backs in. If EV & RV were to be weakened, then I think there'd have to be a corresponding change to make some of the passes complete better against BZ - I'm thinking the quick patterns in particular!

* and possibly the shotgun forms, but they don't seem that reliable against the BZ to me and you have the extra chance of fumbling to contend with as well.

[Dan Slavin]

If EV and RV were weakened, it would remove a staple from the S/D teams who run SL DR DL EV or variations, and would make it much easier to defend. I'd be all for it.

I totally agree that the reverses are too powerful in general and could do with being tweaked. However if they were weakened and everything else left as it is, I'd suggest that it would make all single back formations (apart from Z*) too easy to defend. Basically, reverses seem to me to be the only reliable way to beat the BZ when you haven't got both backs in. If EV & RV were to be weakened, then I think there'd have to be a corresponding change to make some of the passes complete better against BZ - I'm thinking the quick patterns in particular!

* and possibly the shotgun forms, but they don't seem that reliable against the BZ to me and you have the extra chance of fumbling to contend with as well.

Great post.

Jon has also highlighted an issue that I've thought for a while. Effectively a lot of these quibbles are linked. For example, BZ batters most formation/play combos, but S/D EV and S SL works well against it. When you then call passes from S/D you get plenty of deep dumpoffs. To try to limit dumpoffs people call BZ.

[Alex Coombe]

I have always been disapointed that the quick in and quick out are under used and that they could potentially have a vital role in GP in regularly completing against blitzes, and in such a situation RAC yardage wouldn't be out of the question.

I think a nice solution would be to weaken EV/RV agaisnt most plays, especially if it was made to wear out more.
I think DPs should be decreased and primary completions increased on bad matchups, e.g. LO GL ; QL BZ to make up for it.
I find BZ to be very consistent, and only burnt by specific plays as a general rule, EV RW LT etc, you just don't get a play like SW RO CW etc bust out against BZ for a 40 yard run, I think a possible improvement would be the addition of an extra routine for running plays to breakout (with the exception of plays like QS and maybe RC etc), and this breakout ability to be more likely against blitzes like BZ for normal plays not just RN DT etc

Yes I think the problem is too many of these 'faults' are linked and that would make it harder for them to be tweaked,
e.g. if you reduce DPs then passing teams would have quite low completions unless this is compensated for elsewhere
e.g. if you reduce EV's ability the Blitzes are less likely to get burnt from H S C D forms etc, DT doesn't seem to do much for me, not that I've used it much, but it should be great in this situation no?
In the NFL do teams really respond to a heavy blitz with lots of defenders breaking into running lanes and into the backfield by calling slow developing misdirection plays involving unusual ball carriers and multiple handoffs? Of course not, IRL it would be the very last thing I would ever call. I'd go for a SL or a QO, a SC, FO or a DR. EV and RV should be magnificent against soemthing like OV, FD, GL but when youv'e got a corner coming off the edge into the backfield? That's crazy. Gameplan beats the big blitzes in the opposite way to real life IMHO. Just flip the ball to a reciever slipping behind the blitzing players, 5 eligble recievers, 5 players in coverage, everyone is 1:1, any slant, drag or similar route is going to get enough separation for a QB to get the ball in there over the top of the blitz and if he has a step on the defender, lookout!

(a checkdown hotroute audible might be a nice solution to both problems, ie S DS BZ Blitz recognized, pass checked to hot reciever, at a gain of 1 and run for a gain of 4. It would be similar to a blitz pickup routine by a RB)

[Thomas Hennes]

In the NFL do teams really respond to a heavy blitz with lots of defenders breaking into running lanes and into the backfield by calling slow developing misdirection plays involving unusual ball carriers and multiple handoffs? Of course not, IRL it would be the very last thing I would ever call. I'd go for a SL or a QO, a SC, FO or a DR. EV and RV should be magnificent against soemthing like OV, FD, GL but when youv'e got a corner coming off the edge into the backfield? That's crazy. Gameplan beats the big blitzes in the opposite way to real life IMHO. Just flip the ball to a reciever slipping behind the blitzing players, 5 eligble recievers, 5 players in coverage, everyone is 1:1, any slant, drag or similar route is going to get enough separation for a QB to get the ball in there over the top of the blitz and if he has a step on the defender, lookout!

(a checkdown hotroute audible might be a nice solution to both problems, ie S DS BZ Blitz recognized, pass checked to hot reciever, at a gain of 1 and run for a gain of 4. It would be similar to a blitz pickup routine by a RB)

Massively agreed. QI,QO should pretty much beat blitzes anytime (in my mind they're the equivalent of the "hot read", 1-step drop passes), but be very poor against tight coverage (ND) or run/mixed defenses (RD,GL,MD,FD). A QP dump-off would be a great idea as long as it's not too automatic (which is a general problem with dump-offs in GP), meaning that coaches should NOT be able to rely on dump-offs to save their asses when they make bad play-calling decisions. I'd also like to see a WR screen play put in (QS ?). But you're right, quick patterns are under-used when they actually should be the best play selection to beat blitzes.

[Alex Coombe]

QS is taken,

I like HP, dumpoff to a hot reciever, but again rather than a dumpoff perhaps a better idea is just to make some playcalls better against blitzes, rather than bringing in something outside of playcalling to help it balance.
Over my time in AY 1 out of every 8 passes I have called have been PO, a quick pass (with some playaction thrown in there). I haven't used it for a year or two, since I've changed philosophy on offence, but here are some stats for you. I have called it 275 times, for 161 completions (59%) for 3.2 per attempt and 5.5 per completion.
So the percentage is pretty good, at the expense of yardage.
Compare that to SO, which I've called almost as often, 243 att, 126 completions, 6.2 per attempt, 11.9 per completion.
Double the yards for only 7% more completions, and SO gets keyed alot more than PO!

IMX QO is worse, and QI worse still. I can certainly see either of them as a useful play against blitzing (if altered), particularly for a stretch offence which is such a deep threat, you'd find the CBs backing off at times...