Why So Many Draft Trades This Year?

Jets rookies 2022
Photo: USA Today Sports Images

NFL Draft - Guest column by Alex Olbrecht

Two things became very clear as we look back at the opening round of the 2022 NFL draft. First, NFL general managers agreed with the assessment of the QBASE 2.0 model (as seen on footballoutsiders.com) that this year’s quarterback class was weak and not worthy of being drafted very highly. Second, general managers were willing to trade first-round picks at a much higher rate than the past. Eleven first-round picks were traded before the draft, with another nine traded during the draft itself.

The question is why? In short, economic theory can be used to help explain why we observed so many trades in this year’s first round of the NFL draft.

The Coase Theorem states that when conflicting property rights occur, “bargaining between parties involved will lead to an efficient economic outcome regardless of which party is ultimately awarded the initial property rights, as long as the transaction costs associated with bargaining are negligible.” Put into more simple terms: regardless of the initial allocation and order of draft picks, each pick will eventually, through trades, be allocated to the team which values that particular pick the highest. Teams that place a higher value on a pick than the team holding the pick will be willing to offer more for that pick than the value of the pick to the team holding the pick. So we would expect a trade to occur since the team holding pick prefers the compensation to the value of the players available at that pick.  A key point to understand is that for trades to occur, teams need to value a pick differently, and of course value the quality of the player the team can draft with that pick differently.

Analytical models might lead to more consistency with how players are valued by the 32 NFL teams. But evaluating collegiate talent with analytical models is a relatively new technique, compared to the long history of the NFL. The reason why mathematical models haven’t particularly penetrated effectively into the NFL’s player evaluation process is that it hasn’t been until the last 4 to 5 years that data has been available for many positions. In recent years, companies like Pro Football Focus and Sports Info Solutions have created stats and grades for college players at positions such as offensive line and defensive back and thus tried to do some sort of projection and evaluation. Otherwise, analytical models have been limited to players who are tracked with conventional stats. In addition to our quarterback prediction model, Football Outsiders' SackSEER and Playmaker Score systems have attempted to evaluate edge rushers and wide receivers. Fantasy football analysts have been trying to project player performances for years, especially those who are trying to get a leg up on drafting the best rookies.

Measuring player performance in college is very difficult when performance statistics are highly correlated with the play of teammates and the wide variability of the quality of opponents from week to week.  Thus teams are largely left to evaluating various positions based on scouts’ opinions, which we know can be highly inaccurate and biased. Ultimately why this is important for the application of the Coase Theorem is that there is a great deal of variation between the assessment of each team for each player and how valuable that player would be for a particular team. For example, consider offensive left tackle Bob Smith, who most teams view as the best left tackle in the draft. If your team already has an All-Pro left tackle, your team will place less value on that player than another team in desperate need for that position to protect their quarterback while passing.

In some ways, one can think of the quarterback position in the draft as anchoring the value of each pick because the quarterback position is largely viewed as the single most important determinant of winning in the NFL, and thus the price of a draft choice poised to select a premium quarterback prospect is likely to be high. General managers are very clear in their knowledge that trading up in the draft to select a quarterback is going to be very expensive and if that player doesn’t play well, it is likely to be the type of decision to be the cause for a general manager to be fired—both because of the high profile nature of selecting a bust at quarterback and because the amount of draft capital used to acquire that pick will have further weakened the team, making winning more difficult.  However, in a draft with few first-round quality quarterbacks, the issue of variation in the valuation of the picks becomes amplified, the cost of moving up in the draft decreases, and the risk to personal job security of selecting a bust goes down (because the pick won’t be used on a quarterback). In other words, it is a much easier proposition for a general manager to make a trade.

In addition to the valuation of a player that can be picked at a particular position, teams also have financial considerations.  A clear example this past offseason was the explosion in salaries for wide receivers (Gallup, \$62.5M; Williams, \$60M; Robinson, \$46.4M; Godwin, \$60M; Adams, \$141.5M; and Hill, \$120M) which provided a strong incentive for teams with wide receivers with expiring contracts to trade away those players to select wide receivers in this year’s draft.  Since players selected in the draft have artificially lower salaries than they would on the free market, teams have a strong incentive to replace expensive veteran players with cheaper drafted players. If a team evaluates talent well, this can provide significant savings without a corresponding decrease in productivity on the field. Two examples of this type of behavior in this year’s draft were the Tennessee Titans trading away A.J. Brown and the Baltimore Ravens trading away Marquise Brown.

Ultimately evaluating the many positions of players entering the NFL is still very much a judgment-type evaluation by scouts and once financial incentives are considered, the value of each draft pick will differ for each team. The Coase Theorem is pretty clear that once assets are valued differently in a market with low transaction costs, one can expect a reallocation of the assets in question. For the NFL, in a draft with few good quarterback prospects, the risk and costs associated with making a trade in the first round became lower, making trades more likely to occur since the trades are unlikely to be for quarterbacks.  Once you combine that with the imperfect evaluation of talent, general managers were more than willing to deal, guided by the invisible hand of Adam Smith.

Alexandre Olbrecht is a professor of economics at Ramapo College of New Jersey and the Executive Director of the Eastern Economic Association.

43 comments, Last at 16 May 2022, 3:18pm

1Since players selected in…

Since players selected in the draft have artificially lower salaries than they would on the free market, teams have a strong incentive to replace expensive veteran players with cheaper drafted players.

I... don't really agree that's what's happened this year. I mean, 5 teams got rid of high-end WR talent (Packers, Chiefs, Titans, Ravens, Cowboys) and exactly 1 of them drafted a first-round replacement. And it was low-middle first round, too.

I think it's more just that the "WR explosion" is more of a bubble among teams that don't have a top-end QB. Pretty much the only inflated contract by a team with a top end QB is Buffalo. Now, you might say "well, that's obvious, they're the only ones with money" but I think it's more that the contracts have to seem inflated because the agents for the WR don't believe they'll play it out. Because they basically don't.

Bubbles like this seem to happen every 10 years or so, and unless you're Jerry Rice or Larry Fitzgerald, the contracts don't tend to work out. Which means that for those teams, the AAV is pointless (because they don't expect to play it out) and the guarantee/year makes more sense. As opposed to say, Godwin's contract, which isn't inflated at all - it's not even a raise over the franchise tag.

So I don't exactly think that they're swapping high-priced WRs for draft pick WRs (because... they basically didn't) - I think it's more that they look at the market, see it's a bubble, and decide to ride it through with mid-level talent because they've got the QB talent to do it.

2Thus teams are largely left…

Thus teams are largely left to evaluating various positions based on scouts’ opinions, which we know can be highly inaccurate and biased.

They probably still beat their wives, too.

9Don't forget their kids,…

Don't forget their kids, puppies, and kittens.

3It's not a total Coasean…

It's not a total Coasean situation, mainly due to limited time to make draft day deals and limited number of teams.  (Like for NE it kinda seems like they wanted to trade down again and just went with their R2 pick when they couldn't find a partner fast enough).

I would agree the WR trades were more teams dumping the players that took up the most cap space relative to performance that they could handle losing(since they had an elite QB) as well as the fact that they could get something for a player they couldn't keep.  Rodgers and Adams would have take up 26% of the packers cap space next year I think. So it makes sense that they moved when they could find a deal.

4Pic reminds me

A big part of the defense for the trade up for Breece was because they had a good 1st round therefore they could make such a decision.

As if adding 3 rookies (outside the top 3 at that) to a 4-13 team means the rebuild is complete...outside of another RB 😆

Add Sauce = CB room complete

Add Wilson = WR room complete

Add Johnson = Edge room complete

Surely they will all hit and of course everyone else on the team will improve...or at least not regress

Right? Lol

7I don't know who said that...

But the main defense for the trade up is that it's hard to absorb that many draft picks onto your team each year.  You've only got 90 spots initially and ultimately 53.  So the value of a 5th round pick is lower when it's the team's 8th pick that year than when it's 5th pick.  That's particularly true when the team is very young (like the Jets) because, all things being equal, you're going to keep a marginal 2nd or 3rd year player with experience in your system than a late round draft pick.  The more marginal veterans you've got, the more cap space you can save replacing them with marginal draft picks.  So the choices for the Jets were (i) take a guy in the 5th that there's a good chance you cut; (ii) trade the 5th rounder for, say, a fourth in 2023; or (iii) trade the 5th rounder to move up for a guy that you think has first round value.  You can disagree with the valuation of the player, for sure, but as a general strategy either (ii) or (iii) seem like reasonable options.

8Bro, more or less the same argument just in different words lol

Well that and being overly concerned about a singular RB not being able to play every game. Essentially they could go "BPA"

Consolidating assets on a team bereft of talent isn't the strat. They need all the picks they can get, especially premium picks on...premium positions.

They especially shouldn't think their CB room is done after picking CB2 or ED4 in the draft. They letting 3rd year WR Lawrence Cager stop them from doubling up on a more valuable position? Oof. Turrible. I don't care if you valued Ty Chandler or Skylar Thompson as a 1st rounder, that's the whole point, lol, bad process.

Doubling up on edge with Arnold Ebiketie and taking a shot on LB Micah Mcfadden > RB Breece Hall essentially.

16First, a fifth rounder isn't…

First, a fifth rounder isn't a premium pick.  It's most likely a bottom of the roster player or not an NFL player at all.  There are only so many roster spots for marginal players and projects.  If you're talking about never using a second rounder on the clear RB1 with plus receiving skills, then I'll agree to disagree, but that's not really about the trade per se.  That's player eval, not process.  Second, Lawrence Cager -- what are you talking about?  He's a TE conversion project on a reserve/futures contract, a camp body.  Third, they were unlikely to find a CB in the 5th they'd keep over the other guys on their roster.  They've got a bunch of very cheap second or third year players they like as backups (just not as starters).  The CBs picked in the 5th were all projects (like Woolen, who does have potential upside) or backups at best.  Fourth, the Jets actually did double up on edge in the draft in the 4th.

26I didn't say it was...

Part of it was using...38!

Clear RB1? There were plenty that had KW3 (TDN had him 4th fwiw). Either way...it's RB! They supposedly did it to jump Houston. That's silly in todays league. That's literally part of the process. That's the whole point of my pointing to the extremes of Ty Chandler or Skylar Thomson. If you're evaluating them as 1st rounders, that's simply bad process lol and yeah RBs that high for a 4-13 team...is bad process! They just compounded it with the trade up. Like...come on. You literally paid Bell after the Steelers showed even a great RB like him were expendable lol. Let HOU take him and "settle" with a guy that many considered RB1 anyway. IF you're that deadset on getting a RB (again, not where their priority should've been but because of that 1st round they felt, "we can kinda do whatever now")

You said

all things being equal, you're going to keep a marginal 2nd or 3rd year player with experience in your system than a late round draft pick.

2nd/3rd year players shouldn't be prioritized over new picks because of experience. Why would they be for a team that's bad? They aren't bad because they're missing those dudes, they're bad because they don't have talent. And waiting til UDFA isn't the place to do that for every position. The chances that all things are equal are (really) slim and at some point you can't keep beating yourself over the head thinking they'll always develop, and just take another shot at it. Also, semantically, 5th is a mid round along with 3rd and 4ths. 6ths and 7ths are late, but that's neither here nor there for my lexicon.

Their CB room is not stacked. And if they're only trusted to be backups (ok, whatever Ill give it to ya), there's more than one starting CB. Always has been too. It's Sauce (a rookie nonetheless, not guaranteed to hit but lets say he does) and...28 year old Justin Hardee? Or...well 5th rounders (!) Bryce Hall or Michael Carter?

Not that it had to be a specific position. Micah Mcfadden was the pick using the pick they gave up and their LB depth is...CJ Mosley (76th graded LB out of 87) who turns 30 next month and is already expensive before going up next year. Along with...an expiring Quincy Williams (70th)? Those are the only noteworthy ones signed with any sorta relevance. Probably where they should've taken a RB actually. Actually would've been perfect for Tyler Allgeier.

You think doubling up in the 4th is the place to do it? All while arguing the next round is worthless...to a rebuilding team? Oooook. Yeah I think Ebiketie would've been better at 38 and...whoever...in the 5th for a team that left the draft with only 7 picks.

35I agree with you in the…

I agree with you in the general case, but not in this one specific case.  Breece Hall had a BackCAST score of +149.1%, almost 100% over the next running back.  Many scouts did not have Hall as RB1, but he's most likely to turn out to be Jonathan Taylor out of all of the draftees, and the Jets did have a need at the position; Michael Carter the running back seemed solid in his rookie year, but he is not sturdy enough to be a workhorse, and the others were massive problems in the passing game.  The main thing the Jets needed out of this offseason was skill players for their young quarterback, and an interior guy or two to solidify the line in front of Wilson.  The defense can wait, because....

They're not going to do well this year regardless.  Their schedule is brutal (AFC North in a row to open the season, six games against the Bills, the Dolphins and the Pats (the Patriots being the weaker of the strong sisters, when does that happen),  at the Broncos and Packers, and then after week 12 they get the Bears, Lions and Jaguars, maybe some wins there.  They have a good shot at drafting Bryce Young next year, they better know what they have in Wilson.

All that said, I wish they had traded down at some point, and I'm not sure about the trade-up for Johnson.  I would have been fine with them trading down and missing out on Sauce even though I love him as a player.  And as you stated, they need linebackers.  The greatest draft classes tend to be ones where a team had a lot of picks, the 1986 Niners draft being a prime example.  But Breece Hall is going to help a lot with the offense around Zach Wilson, so I'm fine with that pick.

38Breece Hall

From a measurable standpoint, Breece Hall is almost identical to Deebo Samuel. The main difference from a perception perspective is Breece Hall is labeled as a RB that is an adequate receiver while Deebo Samuel is a WR who is an adequate RB. Both positions have had success after the first round of the draft with higher probabilities of elite players in the first two rounds. RB’s make significantly less than WR’s for both high end and mid tier talent. In an era of position differentiation becoming less relevant, is the current salary differential temporary? There are always inefficiencies in the market and with common thought devaluing a better skill player when the goal is to get the best 11 on the field group think has likely went too far. I would anticipate that the forward looking teams (not typically the Jets but they are led by Ravens/49ers personnel now) will exploit this inefficiency until the rest of the league catches up to reverse the trend.

39Well there's that like I said

Michael Carter the running back seemed solid in his rookie year, but he is not sturdy enough to be a workhorse

I would not care if he was a workhorse or not. Which is especially such weird thing to care about coming from a team with a Shanahan disciple. Slap a guy like Allgeier in the rotation and you're golden.

And I just mention those two because that's who the picks turned out to be exactly. Didn't have to be exactly defense even though they can help in indirect ways. Not that using two picks on one guy helps your offense the most.

I don't think a RB (or least Breece specifically) is gonna help you evaluate Wilson as to opposed what could've been had, say Pickens/Raimann and Allgeier. All Breece gonna do is influence them to run earlier and put Wilson in more 3rd and longs. They carried 7 WRs last year and I only see 5 now (less pre-draft) that are guaranteed in Moore, Wilson, Davis, Berrios (does he count as ST tho?) and Mims who they keep sending mixed signals on anyway. Jeff Smith I guess they like but...you could've upgraded with a Pickens or something. But that's still only 6.

This isn't to slander a Breece who I've stated many times was my RB1 as well. It's just the Jets have lot of work to do and RB still wasn't a luxury for them at that point. Double up at OL, WR, essentially anything else other than RB. That position group wasn't the missing link.

He's not Deebo though lol running legit routes is a lot harder than dump offs in the backfield. There a lot more better athletes than Deebo but that doesn't make them Deebo himself. There's also prrooooobably a reason why he didn't run the agility drills (or bench fwiw) like Deebo did. And yeah Jonathan Taylor is great but that's not something you should expect. Or even something to fiend for after seeing his best season...not make the playoffs lol (curse Carson)

5My Econ degree is only ...

a Master's rather than a Ph.D, but I'm better anyway than the author at differentiating a research hypothesis from a research-driven conclusion.

So in past QB-scarce NFL drafts have there been more 1st Round trades than in the just-prior and just-following drafts?

I'm pretty sure you're also maintaining that a sudden and large post-season boost in \$\$\$ for players at 'said' position will boost the # of 1st Round trades. This also would be eminently researchable.

This is a mailed-in article. From an author who I presume did it for free and who has plenty of other being-paid-for responsibilities such that I sure don't begrudge his not taking the time to do any actual research for the article. And on a site understandably happy to get some Friday content during a time of the football year that's as slow as any.

But credentials and 'Theorem' aside, these are all just pure hypotheticals. Such that Aaron et.al. can say "I like this one therefore it's true!" and New et.al. can say "I don't like this one therefore it's false!" (with of course "It's false!" being way more fun)

6as to one research design detail,

Don't position coaches have a lot of input to how linemen etc. are graded? Don't think they should myself, but don't they? Then there are those workouts with the cone drills and all whatnot. Aren't they at least close in importance to what a team's scouts say based just from watching their college play?

13Well, the last time only a…

Well, the last time only a single QB was taken in the first round was 2013 (an all time terrible QB draft.). The next one before was 2000. So...there probably isn't enough data to say either way. The money thing would be hard to research as times when contracts for a position suddenly shot up you'd have to do some weird data joining to find that.  Doable, but probably more work than you would want for a friday fun piece.

Mildly educational at least.

181 less than 2 less than etc.

For the QB thing all you'd have to do is run a simple regression analysis. (seeing as how 1 is less than 2 which is less than 3, etc.) Do the amount of 1st Round trades vary alongside the number of QBs therein taken?

As to the \$\$\$-shooting-up thing, yeah, that'd take some work. Just sayin' here, sayin' here "the 'Ralph' Theorem explains why This! happened in 2021!" is just wrong. Certainly so when you COULD do the research to see if that relationship did indeed exist. Never mind that you properly don't have the time to do that research, that's no excuse for badly overstating things.

29Would probably need to…

Would probably need to account for QB needs to do regression and that gets messy quick. Even if you look at 2 or less taken in the first round you only at 2015 to 2013 and last year for "since 2011".  To small a data set no matter what however.

32how it works

Knot, you look at each draft of course. If 'fewer QBs drafted means more trades', likewise 'more QBs drafted means fewer trades'. Why in the world would you drop the 'more QBs' years from that running?

I also don't see why you'd need to account for QB needs. Do the basic thing first, simple run of 'QBs drafted/trades made' regression. Find no correlation, bid Coase sayonara and find yourself some other Theorem. For next Friday's unpaid article.

40Do the basic thing first,…

Do the basic thing first, simple run of 'QBs drafted/trades made' regression. Find no correlation, bid Coase sayonara

That's... not how regression testing works.

A lack of correlation just means you're failing to reject the null hypothesis - the hypothesis that your original statement was wrong. What effect that has on your actual hypothesis is completely dependent upon what it is - you need an estimate of what you expected the correlation to be, and then a compatibility between the expectation and your prediction tells you to what degree your original hypothesis was in trouble.

In this specific case, the trouble is that the lever arm's small, which is what the previous poster mentioned. There aren't enough drafts with small numbers of QBs drafted to give a reliable estimate there: you're going to be swamped by unconfoundness issues.

I also don't see why you'd need to account for QB needs.

Because the number of years with few highly valued QBs is small, variations in the number of teams who desire a QB (just due to sampling) swamps the overall results. There's also the issue that since QB needs are filled by QBs, the "number of teams who desire a QB" is likely to be lower in years after a "rich QB" class, and higher in years after a "lean QB" class, so your 'basic thing' isn't actually comparing two independent variables.

Basically, your reasoning's reversed: if you do the basic thing and you do find something, it's strong evidence, but doing the basic thing and finding nothing doesn't indicate anything. There's no a priori reason to believe that it's possible to find evidence for a correlation that actually exists in something like sports analytics, because the underlying rules and economics may change faster than the statistics allows for discovery.

15The issue with analyzing…

The issue with analyzing prior occurrences is the rookie pay scale only entered in 2011, and that changed everything. Before that, your 1st round draft pick cost as much or more as a high-value veteran.

To mix metaphors, that's an apocalypse beyond whose event horizon all information is lost.

20True, but still

Yup, that reduces the sample size which reduces Scientific Knowledge Value. And changes nothing with respect to, when you say 'A caused B', even if this is merely Year 2 you still go check Year 1 on that. (and now we are on Year 12 here; so building that data set up)

21Poor Ronald Coase not…

Poor Ronald Coase not getting the love he deserves around here.

25While a standard deviation…

While a standard deviation based upon an n of 2 is mathematically solvable, I would put zero confidence in its accuracy.

10The question is why? In…

The question is why? In short, economic theory can be used to help explain why we observed so many trades in this year’s first round of the NFL draft.

It's also not out of the question that the NFL is being its usual copycat self.   The Rams winning a SB while trading away draft picks like a drunken sailor could well have made a lot of teams think picks weren't nearly as important as they thought.

12I think this was my…

I think this was my hypothesis as well.

I'm sure teams in the past would have traded players like Adams or Tyreke when they were about to hit FA in a year. Just, there weren't the buyers that there are now.

14Before, you didn't usually…

Before, you didn't usually see super stars like Adams or Hill leave their teams before their 30s without some other factor. I think QB will still get paid, we will see when all the current young guys get extensions, but I think it happens.  For other positions, I think teams don't want 1 guy taking up over 10% of the cap, esp given the decline and injury risks. (QB are pretty safe these days, WR...not as much)

11However, in a draft with few…

However, in a draft with few first-round quality quarterbacks, the issue of variation in the valuation of the picks becomes amplified, the cost of moving up in the draft decreases, and the risk to personal job security of selecting a bust goes down (because the pick won’t be used on a quarterback). In other words, it is a much easier proposition for a general manager to make a trade.

Is it, though? The thing about trading up for someone other than a QB is that usually, there are other prospects at that position that the team could reasonably select instead of trading up. So I would put more pressure on a GM who trades up in the first round to select, say, a WR in 2022 to be right. Because if that WR busts, not only is that the waste of a 1st round pick and whatever other assets you gave up, but there is also the likelihood that they could have found a better player at the same position later on.

The rule changes have made QB such a pivot that not only is a miss a multi-year failure, it also makes the other positions fail. A bad pick at WR doesn't do that.

It's hilarious how powerful owners have made QBs. The owners have almost backed themselves into the corner NBA owners have found themselves in, where players openly call the shots up to and including potentially walking out of the playoffs in order to force a change in ownership. I think QBs have enough power and media sexiness to be able to do this.

MLB owners were smart enough to not overly empower any one player position, so they've resisted this problem.

19It's more the nature of the…

It's more the nature of the sport than the owners. Even the best MLB pitchers is every 4-5 games. QB play almost every offensive snap of every game. No MLB position has THAT big an impact.

If Watson can sit an entire year and STILL command the largest contract every and a huge amount in trade even with off field issues we can say the shift has happened.

The rule changes accelerated this since not only do QB have the largest effect, they also have lower injury risk than most other positions(although the trend to runners counteracts that to a degree)

Do goalies have that much power in the NHL/soccer?  Those are the other sports with one position with a much larger impact I can think of.

22Jimmy Haslam

I don't think we can generalize from what Jimmy Haslam does. Particularly in this case. For personal reasons Jimmy has, umm, shall we say, a higher tolerance level with respect to employees' civilly litigable behavior than other human beings do. Including other football owner human beings. (man am I glad I'm not a Browns fan; I found Forrest Gregg a trial, and he weren't nuthin' on the level of Jimmy)

23I care enough to know that, anyway

And while soccer bores the living daylights out of me (kinda fun to play, tho'), you never hear mention of goalies when they talk about who the past or current Great Players were/are. Same with hockey, actually, isn't it?

You hear about great goalies but yeah, scoring is sexier. I think there is also alot more variability. I.e. even the best goalie needs their defense and  there is alot more luck that can happen, so the difference between average and great is smaller so it's not like you can be a goalie away.

Nothing like the Bucs adding brady and going from 7-9 to 11-5 and a SB win so the basically went from out of the playoffs to title contenders. Not only brady, but they were probably a WC out team at best without him.

(Totally agree on soccer, lol. more fun to play than watch)

31You *can* be a goalie away…

You *can* be a goalie away.

Anaheim in 2003 was basically a middling team dragged to game 7 of the NHL Finals by a goalie who was absolutely standing on his head. They lost 6 games in the playoffs. They outshot their opponents twice. They scored 1 or fewer goals 8 times. They won 1-0 three times.

36The 98-99 Sabres finished…

The 98-99 Sabres finished 4th in our own division, then got carried all the way into a Stanly Cup Game 6 tiple OT by Dominick Hasek, who was so dominant they had to allow an illegal goal to finally end it.

37Carey Price dragged the…

Carey Price dragged the Canadiens to the Stanley Cup Finals while the defending champions Lightning considered them a joke; ask Kucherov: https://www.youtube.com/watch?v=oBsOuA5qgFg

The Lightning defeated the Islanders in Game 7 by a score of 1-0.  Kucherov in the interview above claims goalie Vasilevskiy to be their best player; the Islanders themselves only got as far as they did with their goalie tandem of Varlamov and Sorokin.

Another example of how much a goalie can effect a Stanley Cup run is Patrick Roy's 1995-96 season.  The Montreal coach, Mario Tremblay, allowed Roy to stay in a 11-1 rout by the Detroit Red Wings at home, waiting to pull him until midway through the second period.  Roy told the Canadiens President that he had played his last game in Montreal when he came off the ice.  The Canadiens traded him to the Colorado Avalanche.  The Avalanche beat the Red Wings (with their regular season 131 points) in the conference finals and swept Florida to win the Cup that year.

So superstar goalies should ask for more money.  Please don't tell Ilya Sorokin this.

41...and Montreal never…

...and Montreal never recovered. Price carried a mediocre team to the playoffs most years, and look how awful they were this year without him. That's post-Manning Colts bad.

Hasek is the ultimate example, though, and despite being a Bills fan I am NOT a Sabres fan. That team had no business going anywhere in the playoffs, and yet without an illegal goal they probably have their name on the cup.

43I'm an Osgood defender, but…

I'm an Osgood defender, but the Wings were absolutely a better team when they had Hasek. He was a game-changer. Even though he wasn't necessarily a decider in the 7-game series against the Avs, knowing the Wings had Hasek weighed on the Avalanche -- there was always the fear they could dominate a game only to have him steal a win by standing on his head.

28Not with soccer, but the…

Not with soccer, but the save percentage in soccer is low enough that soccer goalies aren't perceived as that good. No one (for limited values of "no one") hates another team's goalie in soccer, and rivalries aren't based upon goaltenders.

This is not true in the NHL. Goalies are much more important in hockey than in soccer. I grew up with the Red Wings, and the fans actively despised Martin Brodeur and were willing to go to war with Quebec regarding Patrick Wah. The Hasek-Brodeur and Hasek-Roy games were absolutely stud goalie rivalries (and you could add Belfour into that mix, too -- https://www.hockey-reference.com/boxscores/200006080NJD.html)

In part this is because soccer will never have this game, which was absolutely controlled by two HOF goalies standing on their heads for 7 periods.
https://www.hockey-reference.com/boxscores/199404270BUF.html

But goalies are kind of their own thing -- like pitchers in baseball. They are separate from position players in a way QBs are not.

30Goalies Powerful?

As others have said, no they aren't, but they're an easy blame target when they don't play well.

34Pretty sure hockey analytics…

Pretty sure hockey analytics says goalies are like RBs:  highly visible and get a lot of press, but you shouldn't spend much of your draft capital or salary cap budget on them.

As I understand it, goal scorers and top defensemen are the positions with the highest impact from an analytics perspective and therefore the recommended players to throw money at.

42What the analytics say is…

What the analytics say is that goal scorers are a more efficient use of resources, not really that goalies are fungible. There's only one Brodeur or Dryden to go around, and there's really no model to predict them to my knowledge - and they may have a 2 to 4 year run of being damn near unbeatable and then return to merely being in the upper echelon after you've paid them a fortune. So they're kinda like QBs in that regard.

33I don’t think so

I don’t think so for two reasons.  One, the amount of capital traded to move up to select a qb is generally more visible—trading high picks to make a really high pick.  Two, the “bustness “ of the pick can become more evident when the pick is a qb, since he is the centrepiece of the offense in a way a WR is not.

DVOA DATABASE: Exclusive Access
• Historical Lookups
• Single Game DVOA by
• Stat Splits by
• Defense
NFL DVOA Database
• Overall DVOA
• Positions
• Line Stats
NCAA Football Stats
In-Season
Fantasy
NCAA
Offseason