SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.
Also, our existing PFR blog rss feed will be redirected to the new site's feed.
Pro-Football-Reference.com » Sports Reference
For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.
ShareThisGuest Post: Quarterbacks and fourth quarter comebacks, Part I
Just above these words, it says "posted by Doug." And it was literally posted by Doug, but the words below the line belong to researcher Scott Kacsmar, who has agreed to write this guest post for us. And we thank him for it.
Forty-seven. The most famous number when it comes to quarterbacks and fourth quarter comebacks is 47. It may also be the most misleading number in NFL history. No matter what source you look at, John Elway is credited with a NFL record 47 comebacks in the 4th quarter. This leads much credence to the "clutchness" of a QB in his career, and is often cited in debates between Elway and other great QBs. But if you research all of his wins, you will have found that he is being credited for comeback wins in a game that ended in an overtime tie, and in several games the Broncos never trailed in the fourth quarter. You cannot come back when there's no deficit to come back from. Dan Marino, always credited with 37 comebacks, has been ranked in 2nd place since retirement. Even DanMarino.com has 37.
Currently, Marino sits 3rd behind Favre (42 is the widely reported number) and Elway (lucky 47). But when the PR staff for the Miami Dolphins exclude wins that the Dolphins never trailed in the 4th quarter, while the Favre and Elway people do not, does that not suggest a serious issue with the validity of these “records”? And why has this been allowed to go on for over a decade?
That's where my research comes in.
In part I, we are going to look at just Elway and Marino’s comebacks. In part II, I will show how several other QBs have had their comebacks tracked and try to create a standard method of crediting comebacks and game-winning drives.
I have taken the time to go through each player's career and get to the bottom of things. As I looked through the games, I used a source for each QB to guide me through how they had arrived at the widely reported numbers.
Marino, from a Dolphins site
Elway, from the HOF website
I will point out the issues with those games, and bring up games missed by those lists. Before starting with Marino, let’s quickly create the definition of a comeback (there will be more detail on this in part II).
For it to be a 4th quarter comeback win, you must:
- Win the game (no ties or losses)
- Take the field with a 1--8 pt deficit (1--7 prior to 1994) and score as an offense (no fumble return TD to win the game)
- It does not have to be the final winning score (hence, that applies to the number of game-winning drives)
Got it? Now it’s time for the data. Note that this includes both postseason and regular season.
Marino is credited with 37 comebacks and 13 game-winning drives (games where the Dolphins did not trail in the 4th quarter and he broke a tie for the win). My data has 51 games, and the boxscores to those games are listed below. An asterisk (*) denotes games that were drives to break a tie and are not comebacks.
*1983-11-06 @ SFO W 20-17 [1] 1983-12-04 @ HOU W 24-17 [2] 1984-11-04 @ NYJ W 31-17 1984-11-11 PHI W 24-23 *1984-12-17 DAL W 28-21 1985-10-06 PIT W 24-20 *1985-10-20 TAM W 41-38 1985-11-10 NYJ W 21-17 1985-12-08 @ GNB W 34-24 *1985-12-16 NWE W 30-27 1986-01-04 CLE W 24-21 1986-11-16 @ BUF W 34-24 *1986-12-14 @ RAM W 37-31 OT 1987-11-01 PIT W 35-24 1987-12-20 WAS W 23-21 1988-10-16 SDG W 31-28 *1988-12-12 CLE W 38-31 *1989-10-08 CLE W 13-10 OT 1989-10-15 @ CIN W 20-13 *1989-10-22 GNB W 23-20 1989-11-19 @ DAL W 17-14 1990-09-09 @ NWE W 27-24 1990-10-07 NYJ W 20-16 1990-12-09 PHI W 23-20 OT 1991-01-05 KAN W 17-16 *1991-09-22 GNB W 16-13 [3] *1991-11-10 NWE W 30-20 1991-11-24 @ CHI W 16-13 OT 1992-09-14 @ CLE W 27-23 1992-09-27 @ SEA W 19-17 1992-10-11 ATL W 21-17 1992-11-22 HOU W 19-16 1992-12-20 NYJ W 19-17 1992-12-27 @ NWE W 16-13 OT 1993-09-05 @ IND W 24-20 1994-09-04 NWE W 39-35 1994-10-16 RAI W 20-17 OT 1994-11-06 IND W 22-21 1994-11-27 @ NYJ W 28-24 1995-10-01 @ CIN W 26-23 *1995-11-05 @ SDG W 24-14 1995-12-03 ATL W 21-20 *1996-11-17 @ HOU W 23-20 [4] 1997-09-07 TEN W 16-13 OT *1997-10-05 KAN W 17-14 *1997-12-07 DET W 33-30 1998-10-25 NWE W 12- 9 OT *1999-01-02 BUF W 24-17 1999-10-10 @ IND W 34-31 1999-12-19 SDG W 12- 9 2000-01-09 @ SEA W 20-17
There are four issues with these games (the ones with a bracketed number) that I have resolved.
[1] It is not known what 13 games a few sources have listed as the games Marino led a game-winning drive to break a tie, but I’m guessing they are missing one from his rookie season against the 49ers (source). The Dolphins led 17-14 to start the 4th before Joe Montana drove the 49ers to a game-tying field goal. Marino led the game-winning field goal drive for the Dolphins, picking up yardage on a Ronnie Lott pass interference penalty. This makes Marino +1 in overall 4th quarter wins (51).
[2] The dolphinsinfo.com source misses Marino’s very first comeback, also from his rookie year. At Houston on 12/4/83, the Dolphins trailed 17-10 starting the 4th. Marino threw a TD to Tony Nathan to tie the game at 17. Later he led an 82 yd drive for the winning TD in a 24-17 win (source). Marino was injured and left the game the play before the Nathan TD. But by virtue of his TD to Nathan and being out there for most of the game-winning drive, this is most definitely a 4th quarter comeback, the first of Marino's career. So Marino is +1 in both comebacks (38) and overall GW drives (51).
[3] 9/22/1991 vs. Green Bay - the Dolphins started the 4th quarter trailing 13-6. The Miami offense had been rather impotent to this point. NT Chuck Klingbeil recovered a Don Majkowski fumble for a TD to tie the game at 13. So this cannot be a comeback since Marino never did anything while trailing. Marino completed a 40 yard pass to Duper to set up the winning field goal for a 16-13 win. This is a game-winning drive, not a comeback. So subtract a comeback (that'd keep him at 37).
[4] 11/17/1996 vs. Houston - Similarly, the Dolphins trailed 17-13 to start the 4th quarter. Zach Thomas scored on an interception return to take a 20-17 lead. This cannot be a Marino comeback. The Oilers tied the game on a field goal at 20-20. Marino then led a game-winning field goal drive for a 23-20 win. This is a game-winning drive, not a comeback. Subtract it from the total (36).
Marino results: Marino had 51 overall wins decided in the 4th quarter/overtime, and 36 of them are comebacks. This is one less than what he's usually credited with.
Elway is credited with 47 comebacks. According to the HOF article and in the Denver media guide, "Elway chalked up a record 47 fourth quarter come-from-behind comebacks during his pro career." Come-from-behind? Not quite. Here is my list of 50 Elway games, again with asterisks pointing out games that were not comebacks.
1983-12-11 BAL W 21-19 1984-11-04 NWE W 26-19 1984-11-11 @ SDG W 16-13 *1984-12-09 SDG W 16-13 1985-09-22 @ ATL W 44-28 *1985-10-20 SEA W 13-10 OT [1] 1985-11-11 SFO W 17-16 1985-11-17 SDG W 30-24 OT 1985-12-01 @ PIT W 31-23 1985-12-14 KAN W 14-13 1985-12-20 @ SEA W 27-24 1986-09-07 RAI W 38-36 1987-01-11 @ CLE W 23-20 *1987-09-20 @ GNB T 17-17 OT [4] 1987-11-16 CHI W 31-29 1987-12-06 NWE W 31-20 *1988-01-17 CLE W 38-33 1988-10-09 @ SFO W 16-13 OT 1989-10-08 SDG W 16-10 1989-10-22 @ SEA W 24-21 OT *1989-11-12 @ KAN W 16-13 1990-01-07 PIT W 24-23 1990-09-17 KAN W 24-23 *1990-09-23 SEA W 34-31 OT [2] *1990-10-21 @ IND W 27-17 *1991-10-20 KAN W 19-16 *1991-10-27 @ NWE W 9- 6 *1991-12-08 @ CLE W 17- 7 1991-12-15 PHO W 24-19 1992-01-04 HOU W 26-24 1992-09-06 RAI W 17-13 1992-10-04 KAN W 20-19 1992-10-18 HOU W 27-21 1993-12-12 KAN W 27-21 1994-10-23 @ SDG W 20-15 *1994-11-13 SEA W 17-10 [3] 1994-11-20 ATL W 32-28 *1995-09-17 WAS W 38-31 *1995-11-19 SDG W 30-27 1995-12-24 @ OAK W 31-28 1996-09-15 TAM W 27-23 1996-10-20 BAL W 45-34 1996-11-04 @ OAK W 22-21 1996-11-24 @ MIN W 21-17 *1997-10-26 @ BUF W 23-20 OT *1997-11-02 SEA W 30-27 1998-01-04 @ KAN W 14-10 *1998-01-25 @ GNB W 31-24 1998-11-01 @ CIN W 33-26 1998-12-06 KAN W 35-31
Again, several issues here. First, the Broncos' PR people must have fallen asleep while looking at games against the Seattle Seahawks. There were three Seattle games where Elway should be credited with a game-winning drive (but no comebacks since they never trailed).
[1] 10/20/85 vs. Seattle – Denver led 10-7 to start the 4th quarter. Seattle forced OT with a field goal. After the teams traded punts twice, Dave Krieg threw an interception that put Denver at the Seahawks' 15 yard line. The offense, with Elway, came out and ran 3 plays for 8 yards. They kicked a 24 yd FG for the 13-10 OT win. Hardly the stuff of legends, but it still counts as a game-winning drive.
[2] 9/23/90 vs. Seattle – Denver led 28-24 to start the 4th. After adding a field goal to the lead, Seattle forces overtime with a TD for a 31-31 tie. In OT, Elway completes three passes and Bobby Humphrey ran 26 yards on a draw on a 66 yard drive that led to the winning field goal.
[3] 11/13/94 vs. Seattle – Denver led 10-3 to start the 4th quarter, only to see the Seahawks tie the game on a TD run. Denver answered with the winning TD drive, capped by a Leonard Russell 11 yard TD run. On the 9-play, 80 yard winning drive, Elway completed all five of his passes.
None of the three are comebacks, but they are game-winning drives. I was thinking the reason Denver people didn't list the first two is because they have nothing to do with the 4th quarter; they were OT drives. The third one was not, and they just flat out missed it.
But if you look closely at that Denver/HOF link with the 47 games, you'll find this gem:
"Oct. 26, 1997 at Buffalo — Directs 9-play, 43-yards drive in 4:47 during overtime to set up a 33-yard Jason Elam field goal with 1:56 remaining on the clock, giving Denver a 23-20 win."
That’s right; it is the exact same situation as the first two games they missed. This is why my Elway data includes 50 games (the 49 wins and the tie). Of those 50 games, 15 of them saw Elway lead a game-winning drive without ever trailing in the 4th quarter. Subtract these 15 from the comeback total, and that makes it 35 comebacks.
Some people have mentioned backup QB Gary Kubiak filling in for an injured Elway and finishing off a winning drive, thinking that Elway should not get credit. This was on 12/20/85 vs. Seattle (something about Elway and Seattle). That thinking is wrong. Twice, Elway led game-tying touchdown drives in the 4th quarter when Denver trailed by 7. With the game tied, he completed a 27 yard pass, and then was knocked out of the game. Kubiak scrambled to finish off the drive and they won on a field goal. Elway deserves credit for this one, one of his finest comebacks and NFL games period.
[4] Finally there's the issue of the 17-17 tie against Green Bay in 1987. Granted, the Broncos were down 17-10 and he led an 18-play drive to tie the game at 17. But the next 20+ minutes of the game were scoreless and the game ended in a tie. While this technically can be seen as a comeback, there's no win attached to it. Call it a tying comeback, a special situation, but do not call it a comeback win. Subtract this one too.
Elway results: Elway had 49 overall wins decided in the 4th quarter/overtime, and 34 of them are comebacks, and another game was a comeback that produced a tie. This is 13 fewer than he gets credit for, and the overall number of drives is fewer than Marino's.
Even if you count the tie, it's still 51 to 50 in favor of Marino. If you count the tie as a comeback, it’s still 36 to 35 in favor of Marino for comebacks. No matter what you wish to call them, Marino has more than Elway, and deserves to be recognized for it.
Next time we’ll look at some other QBs and how their comebacks have been tracked, not to mention how Elway may actually be only third all-time instead of the clear #1 position he has held for over a decade.
In the mean time, if you're interested in more studies of 4th quarter comebacks, see Jason McKinley's article at Football Outsiders, and Clark Heins' research, which you can read about here.
Feel free to send any special questions or comments about this to me at smk_42@yahoo.com.
Editor's note: here is a direct link to Part II of this article.
This entry was posted on Thursday, August 6th, 2009 at 5:02 am and is filed under History. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
And here's an official source http://www.miamidolphins.com/newsite/history/danmarinotribute/danmarinotribute.asp for that 50 number Marino is less widely known for.
Very interesting. I notice on Marino's list only four of his tie-breaker/comeback wins were playoff games - vs. Cleveland in 1985, Kansas City in 1990, Buffalo in 1998, and Seattle in 1999.
Thanks for the reference to the Football Outsiders article. I enjoyed the use of statistical info that I have familiarity with.
Still (and I know this from my field, psychology), the problem of operational definitions and confounding variables diminishes the analysis. But still, it is better to "do the math (stats)" and see where it goes than depending upon impressions and myth.
Finally, it is good to see your numerical analysis to dispel some of the mythology of this area. Numbers get thrown around without them being evaluated carefully. As always, good work in an area that makes this site so special.
Hmm... I never really liked the "4th-quater-comeback-metric". It seems awfully arbitrary. If a team falls behind 31-0 in the first half, and the QB then throws 5 td's in the third - he'll get no credit for the remarkable comeback.
"The Hidden Game of Football" has a metric called something like win-probability. They use it to figure out what's better; to kick the field goal or go for it. Couldn't this be used to figure out how many big positive swings in VP a QB has had during his career? Obviously divided be the time that was left in the game...
Did that make any sense?
Great Stuff, Scott. Post here any time.
Really interesting stuff, Scott. I never thought much about the issue and I'm surprised that so many people cite these unofficial numbers as official.
Great stuff, but I'd be even more interested in something like "comeback percentage."
I got tired of all the sources this offseason noting that Brett Favre had 42 come-from-behind wins (in the same way people tend to forget his awful games while elevating him to sainthood for his great games) and I wondered how many "come-from-behind losses" he had and how frequently he succeeded.
In other words, given the chance to come from behind in the 4th quarter, what is a QB's success rate? That, to me, means more than raw numbers.
DDF, that sounds like a good idea. Last year Doug pointed out a site (click my name) that calculates win comeback probabilities. Unfortunately, their database only goes back to 2000. It looks like their top comeback since then was when the Jets beat the Dolphins on Monday night in 2000. They get a 100 comeback rating for that game.
So it would be cool to sum the comeback rating for every game that a QB wins, to give them a career total.
One thing I hate about QB comebacks as a stat is that it totally ignores the responsibility the QB might have for being down in the first place. Jake Plummer is a good example of this. Why should we care how often a QB led his team to a comeback if he was the one who threw INTs during the first 3 quarters?
OK, I looked at the comeback ratings from the Advanced NFL Stats website. I took every comeback with a rating of 20 or greater between 2000 and 2008.
Here are the guys that had 4 or more comebacks:
Manning 10
Brady 7
Delhomme 6
Testaverde 6
Garcia 5
Brooks 4
Bulger 4
Couch 4
Manning, Eli 4
McNair 4
Rivers 4
Here are the top 21 in cumulative comeback ratings:
Manning 661.6
Rivers 333.3
Brady 324.9
Garcia 308.3
Testaverde 265
Bulger 258.3
Delhomme 218.3
Brooks 216.6
Couch 203.3
Grossman 200
Matthews 200
McNair 195
Manning, Eli 178.3
Hasselbeck 175
Bledsoe 170
Dilfer 153.3
Garrard 133.3
McCown 133.3
Plummer 133.3
Romo 133.3
Simms 133.3
Excellent work and can't wait for the next post. While Elway may be an all-time great QB, I've always felt that he was somewhat overrated. The way people talk about him(so many call him the greatest ever) seems out of whack with reality. And like Favre, I can't help but wonder if Elway's play from quarters 1-3 helped create those situations in the 4th/OT where he had to win it. I don't base that on numbers...that's based on the (possibly)lyin' eyes belonging to myself and others.
I think plunging Sammy Winder into the line for three quarters helped to create those situations, so Dan Reeves should get a lot of credit for Elway's comeback totals. Take 10 more off Elway's total for that.
I think Zippy hit the nail on the head. When I think of Elway's 4th-quarter heroics, I'm always reminded of Reeves' offensive philosophy of the day- hyper-conservative for 3 quarters, then turn Elway loose in the 4th and let him bail you out. That's why Mike Shanahan got fired (because he and Elway were cooking up and calling plays behind Reeves' back, because Reeves' gameplan was too conservative an ineffectual).
Also, I'm one of those who has argued that Elway was the best ever, but it has nothing to do with his "comebacks" and everything to do with the talent level around him. Several QBs have done more, but no QB has ever done more with less. Elway played 12 years before he ever had a pro bowl WR on his roster. He had one pro bowl RB during that span, and for all the talk about how bad the running games Marino had to work with were, if you subtracted out Elway's rushing yardage (and added Marino's lost yardage back to the Dolphins), the Broncos and Dolphins essentially averaged the same amount of yards per season. He had fewer pro bowl offensive linemen than the other greats, too. Despite really being a one-man offense, Elway still managed to make 3 SBs in 4 years in the late '80s. I honestly believe if you'd put Elway in San Francisco and Montana in Denver, that Elway would have as many rings as Montana had but Montana wouldn't have as many SB appearances as Elway had.
Re: #2 Mike Daly
Yes, Marino had 4 in the postseason, but performance-wise he could have been 6/6, always coming through when given a chance in a close playoff game.
In his first playoff game against the Seahawks his rookie year, the Dolphins trailed 17-13. A Dave Krieg INT sets them up with good field position, Marino completes a 14 yd pass to Nat Moore and they score the go-ahead TD. Dave Krieg answers back with a TD, then Miami return man Fulton Walker fumbles the ball. Seattle adds a FG (27-20 lead), and Walker again fumbles the ball. Marino never got another chance.
In 1994 Marino was having a great playoff run. They led 21-6 in San Diego at halftime. The offense would only run 16 plays in the second half as San Diego came back to take a 22-21 lead. Marino led a late FG drive, but the 48 yard FG was no good with 0:01 left.
Re: #7 Jason W
Be sure to read tomorrow's entry as I look at the ways I think you should analyze comebacks. Getting a percentage (or success rate) is definitely important. It's also very hard to accurately do for QB's that started their careers before the GameCenter/play-by-play era.
And in mentioning Favre, I have him with 52 failed comeback attempts and that's only back to 1995. I'm sure the numbers go up when adding his first 3 seasons as a starter, so he really was not that great at delivering comebacks or GW drives.
Re: #9 Bowl Game Anomaly
A future goal is to look at comebacks that have play-by-play and see how the QB played before that point. Nearly half the games in the NFL are "close" in the 4th quarter, so it's not fair to say all comebacks are products of poor play in the first three. Look at Super Bowl 43 for example. Roethlisberger gets credit for the comeback win, but it's not like he was ever playing bad in the game.
I think the games where a team has to come back from a 2-score deficit in the 4th may result in more poor QB play thru 3 quarters as opposed to games where it's just one drive needing one score. That's my hypothesis.
good stuff i am finally happy someone knows something about comebacks...elway put himself in 4 quarters situation...which lead to believe his qb rating from 1-3 quarters were horrible!!!!!!!!
What would be cool, but pretty timely, would be to have the number of 'failed' comebacks, and thus, the % of comebacks wins (% of possible opportunities that is).
Elway was one of the most exciting QBs to watch, though I don't think he was a particularly gifted passer. His 98 yard drive against the Browns in the last 5 minutes or regulation in the 86 playoffs... That was amazing. 3rd and 18, the crowd is deafening, Elway is covered in mud, and the snap bounces of the receiver in motion's hip. Elway recovers it and throws for the long first down, keeping the drive alive. Of course, Elway took the sack that put them in that situation in the first place...
Something weird I noticed about that game, Elway completed 22 passes, but nobody on his team had more than 3 receptions.
I'm not a fan of these "clutch" sort of stats because I don't really believe in them. There's just too many other factors for me. Take a thousand 4th quarters where the Pats trail the Lions by a point -- is it really surprising when the Pats "come back" in most of them? I don't think so -- It's to be expected given the teams. If you reverse the situation, then it's pretty darn amazing when the Lions pull one out. So I expect QBs who've played on strong teams will end up with a higher % of 4th quarter comebacks. Not to take anything away from em -- Any team that comes back must be doing something right. I just think it's much easier to be doing something right in the 4th quarter when you're the better team going in to the game... and guys who've played on strong teams will be much more likely to be on the better team going into the game.
I was never so impressed with this stat... wouldn't you prefer your quarterback to have his team in the lead after 3 quarters? Ideally?
To clarify -- I love the trivia aspect of this post! I just don't buy that it's indicative that one QB is more "clutch" than another. Very cool, and good job on the research
Andy, I didn't include it in the article because they are estimates and I like to deal with facts, but I did come up with the total number of opportunities for Marino and Elway.
Note that these include more than just comeback opportunities, as games with a tie are included.
Marino - I found 51 games where the Dolphins lost, and Marino had the ball in the 4th QT/OT, either tied or down by 1-8 pts at least once. Now I can only verify 48 of these games. There were an additional 3 where I couldn't verify if he did or didn't have the ball (the score of the game keeps the window of opportunity open in these cases). Add in the 51 successes and for Marino, it's a min of 99, max of 102.
Elway - I found 57 games where the Broncos lost, and Elway had the ball in the 4th QT/OT, either tied or down by 1-8 pts at least once. I can only verify 52 of these games. Add in the 50 games here, and for Elway, it's a min of 102, max of 107.
So I can reasonably conclude that Marino had 2 more wins decided in the 4th QT/OT in anywhere from 0-8 more opportunities.
I can also tell you that Marino, in the 51 suggested losses where he had the ball down by one score in the 4th, had his team ahead in the 4th QT 24/51 (47.1%) times. Elway, in the same situation, had his team ahead 14/57 (24.6%).
Correction: Marino had 2 more wins in anywhere from 0-8 FEWER opportunities.
Where was roethlisberger on the list? He has 18 4th quarter comebacks in 52 career games.
Go to part II
Ok. This guy is a total hack. He's not data-mining, but rather absolutely lying or misleading.
He lists Oct 20th, 1985 as a game that should not have been counted as a 4th quarter comeback for John Elway.
Well guess what, IT NEVER WAS TO BEGIN WITH!
Using his HOF link, which he posted himself, you can clearly see the game was never mentioned in the official list.
[editor's note: he never said it was listed at the HoF link. In fact, he specifically said it wasn't. Read it again.]
I've been saying this for years. We old timers knew that Marino was the real King of the Comeback. I suspect they pushed Elway because Marino was already so marketable, he didn't need it. Either way they were both impressive.
The site also exposes Favre for being the fraud he is (Fraud=Hype./ I've been saying that for years too.
Great article guys. You're the best there is.
Thank you,
Tony
Aka
FtballGuru
FAVRESOVERRATED
300Interceptionclub
Yadda
Yadda
Yadda
Great article about two of my all-time favorite quarterbacks. The only issue I have with your list are the games where Marino and Elway left with injuries. I don't think they should get credit for those, much as that may suck. With all the variables that can happen on any play (INT, fumbles, sacks) you can't say for sure that they would have finished off those drives, and therefore, I think it was correct to leave those off their respective lists. Just MO, tho. Nicely researched and written article. Thanks.
Led, I know what you mean but in each case there was only one game like that for Marino and Elway. And it still would not effect the number of comebacks they had as they each completed a TD drive (Elway had two) before getting hurt in those games. It would change the number of game-winning drives had you decided to credit the backup, but the comebacks wouldn't change.
Is there any source that is counting comebacks for this season 2009?
Scott,
Very interesting article.
I've often wondered how many times the great comeback quarterbacks had someone else come back on them. You didn't address that directly in your article, but in a later post, you say you were able to verify 48 games the Dolphins lost in which Marino had the ball in the 4th or OT at a tie or 1-score deficit, and in 24 of these, Marino was ahead at some point in the 4th. You say you were able to verify 52 games the Broncos lost in which Elway had the ball in the 4th or OT, tied or behind by one score, and in only 14 of them Elway was ahead at some point in the fourth.
What that really means is that teams came back on Marino 10 more times than they came back on Elway. The games you mention in which the Dolphins or Broncos were ahead at some point in the 4th, 24 for Marino, 14 for Elway, does not only include games like the famous Chiefs-Broncos (10-17-1994) game, in which Elway came back on the Chiefs, and then the Chiefs came back on him, it includes games in which the Dolphins or Broncos were LEADING at the beginning of the 4th, and lost. And the rest of the games you verified, 24 for Marino, 38 for Elway, do not only include games where they had the whole 4th quarter to get one score, and couldn't get it done, they also include games where they had to score on every 4th quarter possession just to get close.
Also, according to pro-football-reference.com, Elway lost 89 games (including playoffs). the 52 close Bronco games you verified are 58% of those losses. Marino lost 103 (again, including playoffs). The 48 Dolphin games you verified are only 47% of his losses. So while Elway's losses were usually very close, Marino's losses tended to be a bit more lopsided.
I have to take issue with the way you added what you say is Marino's first game-winning drive. You say "It is not known what 13 games a few sources have listed as the games Marino led a game-winning drive to break a tie, but I’m guessing they are missing one from his rookie season against the 49ers." First of all, you don't name these "few sources." They may not be counting the same games, they may not be counting the same way, or they may be flat-out wrong. Second, you can't use a guess to add to his total.
Your numbers for Marino come from a Dolphins website (among others, apparently), while your numbers for Elway come from the Hall of Fame website, a neutral source. I think that speaks for itself.
Brad, the Elway link on the HOF website is not written by the HOF. Notice it says the source is from the Denver Broncos themselves on that page, so it's no more neutral than the Miami one.
I did not make any guess in adding games to his total. The facts and sources are there. All games count.
As far as Elway's losses being closer, well we know Marino played, on average, with worse defenses than Elway did in their careers. I count 10 more starts in Marino's career where his team allowed 30+ points (a difficult situation to win a game under) compared to Elway's. Without doing the math, those 10 games alone should get the percentages closer. Also I see 4 more 40+ point games in Marino's career compared to Elway's. Considering Marino has 41 games where they allowed 34+ and Elway had 24 such games, it's safe to say the average 30+ pt game for Marino was even harder to overcome than it was for Elway.
I think no matter how you look at it, Marino did a better job of having his team ahead in the 4th quarter than Elway did. Whether the rest of the team finished as well, that's another debate.