Pelton mail: Curry more valuable to Golden State than Durant

Kevin Durant: 2014 MVP, 2017 Finals MVP. Stephen Curry: 2015 MVP, 2016 MVP. Kyle Terada/USA TODAY Sports

This week's mailbag features your questions on the impact of Kevin Durant, Stephen Curry, Steve Kerr and more.

You can tweet your questions using the hashtag #peltonmailbag or email them to peltonmailbag@gmail.com.

Ben's tweet was in reply to one of mine about Steve Kerr's post-championship appearance on the Lowe Post earlier this week. Discussing the topic of the best players in the league, Kerr told Lowe, "LeBron's the best player in basketball, and KD [Kevin Durant] is right there with him."

When Lowe followed up about mentioning Durant and not Stephen Curry, Kerr explained the choice.

"Steph is one of the great players in the world, a two-time MVP, incredibly dominant in terms of the way he impacts a game. But you're talking about Kevin Durant at 6-foot-11 protecting the rim, scoring 35 a game in the Finals and getting to the line, doing whatever he wants to do. There's a different level of impact that is actually possible for Steph and for KD.

"Steph is my size. He's literally exactly my size, so what he's capable of is all based on his amazing skill. But he can't do anything about being 6-foot-3 and 175. In the playoffs especially, when you're having to protect the rim and do so many different things, there has to be a level of physicality that factors in when you're having these conversations about who's the best player in the league.

"That's why LeBron is there. That's why Kawhi [Leonard] is there. That's why KD is there: that combination of skill and physical, sheer force."

Kerr knows both players better than almost all of us outside the Golden State Warriors ever possibly could, and he has forgotten more about basketball than I'll ever know. So it's with great trepidation that I tell you I disagree with his opinion about the relative value of Curry and Durant.

Everything Kerr says about Durant's ability to control a game at both ends is true and was on full display during Game 2 of the NBA Finals when Durant played center at times during the second half to match up with the Cleveland Cavaliers' small lineups. So my dispute is mostly a matter of the difference between the two players on offense.

Durant was the more efficient scorer this season, and the two players used a similar percentage of the Warriors' plays. Yet I still think Curry is the much more valuable offensive player, starting with the value he provides as a passer. Curry also tends to create more of his own shots: 56 percent of Durant's playoff field goals were assisted, per Basketball-Reference.com, as compared to 38 percent of Curry's.

Then there's the crucial matter of Curry's gravity. Gravity is our term for the attention an offensive player attracts from the defense. (See here for a full explanation of the concept.) Because of his unparalleled 3-point shooting, Curry has more gravity than anyone else in the league, as was evident during Game 1 of the Finals when Golden State got two uncontested dunks in transition (one by Durant) because Cleveland defenders ran to Curry at the 3-point line.

While Curry's gravity isn't always that obvious, his shooting helps open up the floor for teammates throughout a game. That's the best explanation for why Curry continues to have far greater impact on the Warriors' team success than Durant.

During the regular season, Golden State outscored opponents by 14.4 points per 100 possessions with Curry on the floor and not Durant, according to NBA.com/Stats. Reverse that and the Warriors drop to 6.2 points per 100 possessions better with Durant and not Curry. (They were plus-19.2 points per 100 possessions with both superstars on the court together.)

That comparison is a little unfair because Curry played more with the other starters in Durant's absence due to injury, whereas Durant's minutes without Curry were mostly alongside reserves. In the playoffs, after Durant returned from injury in Game 4 against the Portland Trail Blazers, the two players played with similar teammates. Yet Golden State was plus-14.9 per 100 possessions with just Curry and outscored (minus-2.5 per 100 possessions) with just Durant.

As a result of that performance, ESPN's real plus-minus (RPM) saw Curry as far more valuable than Durant this season, attributing him nearly three points per 100 possessions more impact on the Warriors' offensive rating and rating him 1.7 points per 100 possessions more valuable overall.

It's possible the effect this year was a function of Golden State trying to fit in Durant and he'll have more impact next season. It's also possible that Durant's size does make him more valuable against the very best competition, as in the NBA Finals, even though Curry is superior against lesser foes. (I'm not willing to draw that conclusion based on five games, particularly when Curry was outstanding in his own right.) For now, if I'm assessing which player does more to help the Warriors win, I'm going with Curry.

"I think I follow too much #warriorstwitter, but I'm curious as to your thoughts. Steve Kerr is obviously an amazing coach; even if you throw out the results, it's clear what he's done for the culture, the team's collective confidence and psyche, the beautiful offense, and the delicate task of integrating KD into a team that had already won a chip and had the greatest regular season ever. At the same time, he's a human being -- and thus, by definition, not perfect -- and it's easy to forget that he's only been at this for three years.

"What do you think of the criticisms, particularly from Dubs fans/bloggers/obsessives, surrounding his rotations, his reluctance to ride his stars, and his antipathy toward PnR/iso basketball? Are these actual problems? Do the negatives of his philosophy -- both schematically and the Strength in Numbers stuff -- outweigh the positives? Or are people just looking for *something* to blame every time the Warriors are less than perfect or outside factors (suspensions, injuries) keep them from the mountaintop?"

-- Michael Smith

Speaking of Kerr -- I mean, obviously, the positives outweigh the negatives. What I think I've settled on over the course of this year's playoff run is that there are better coaches to win a single game, and maybe even a series, but not for a full season.

Because of his obsessive focus on the bigger picture, and keeping guys involved, I do think Kerr tends to leave some points on the table at times in a way that Mike Brown didn't when he was running the team during the first three rounds of the playoffs. If Kerr's at the helm, it's very possible Golden State's undefeated postseason ends in Game 3 of the first round at Portland, when Brown extended the minutes of his starters to get a win that at the time seemed to mean only the difference between sweeping the series and winning in five.

Yet, with the exception of last year's NBA Finals -- and barely so -- Kerr is completely justified in his position that the Warriors can win without chasing those points, and I think that perspective is invaluable over the course of a long season.

Game 3 of the NBA Finals might have been the ultimate test of Kerr's philosophy. The Cavaliers were ahead and pushing the minutes for Kyrie Irving and LeBron James, yet Kerr remained resolute in his belief that they would eventually wear down. He stuck to his game plan, got his team to buy in and got the W without sacrificing anything in the long term.

I totally understand how frustrating Kerr's philosophy can be for Golden State fans -- I feel it myself as a neutral observer -- but we shouldn't mistake frustrating for bad. It's only because of how good Kerr has made the Warriors that any small stumble seems so frustrating.

"Why not eliminate the max individual contract to spread out the star players? There must be a good reason it is never brought up."

-- Pat

Well, there's not really a ton of incentive for either side to make it an important factor in negotiations on the NBA's collective bargaining agreement (CBA).

Owners pushed for the maximum salary -- it was a major reason for the 1998 lockout -- and not unreasonably feared that a star could demand exorbitant sums from a team with his Bird rights. Kevin Garnett's extension was the big flashpoint for a maximum salary. When it kicked in, after the lockout, he alone made 46.7 percent of the Minnesota Timberwolves' salary cap, far higher than the current 35 percent maximum salary.

On the players' side, while Michele Roberts has spoken out about her issues with a salary limit and star players wielding more power in the players' association than they have in some time, the issue remains that each additional dollar stars make ultimately comes out of the pockets of their lower-paid peers. And with many more average NBA players than stars, that's a tough sell.

Raising the maximum salary limits is one way to get the benefit you describe in terms of forcing stars to sacrifice more money to team up, and indeed by tying maxes to the actual cap instead of a slightly smaller amount, the new CBA does somewhat increase max salaries.

Some context here: On this week's episode of the Seattle sports and food podcast that I host with my brother, the Fabulous Peltoncast, I described a local burger chain as having replacement-level burgers.

My brother asked which NBA player best epitomized replacement-level play, and the best answer I could come up with was Tim Thomas, a reference to when Bill Simmons came up with "value over Tim Thomas," and I subsequently found under the version of my wins above replacement player (WARP) stat I was using at the time that Thomas had rated exactly at replacement level two seasons running. (Subsequently, when I added credit for floor spacing to WARP, Thomas' rating improved.)

Of course, Thomas has been retired for several years, so who's the current replacement-level leader? I used my combined rating that factors both WARP and RPM, looking at which players who had played at least 1,000 minutes each of the last five seasons totaled the smallest squared differences from replacement level:

Jeff Green is in the mix but ranks sixth. I don't know how Avery Bradley ended up in there, either, but Dante Cunningham is the narrow winner over Luc Richard Mbah a Moute as the most replacement level.