Should a 7-2 win be more valuable than a 5-4 one? That’s the debate and something I would like to discuss. I’m advocating for a change from 9Cat weekly scoring to W/L weekly scoring. Read on... [Ed: This was written right when the NBA shut down, so the final standings and some numbers are slightly off as Fantrax hadn't updated WK21 fully yet]
First, let’s set some terminology. Moving forward, our current system of rewarding a WLT for each category per week will be referred to as “9Cat.” The alternate method, awarding a single win loss or tie per week will be referred to as “W/L.”
[ 9Cat vs W/L | W/L Standings | W/L 2019-20 Season Compile ]
Image Explanation: On the left side is our actual standings from the twenty-one weeks of the regular season. On the right side is if teams were awarded a single win loss or tie instead. “Rank +/-“ and “% +/-“ indicates the change going from 9Cat to W/L in rankings for each individual team. Note: Seeding for the W/L calculations isn’t exact as I didn’t go through to figure out H2H matchups, so the higher seed just went to the team higher up on the alphabet. Precise, I know...
Analysis
• Switching from 9Cat to W/L would not have affected nineteen of our thirty teams in the final standings. Those nineteen teams would have had the exact same power rank — and seeding — with either method.
However, five teams would have gone up a spot or two in the standings — GSW, PHI, MIN, UTA, LAC — with GSW and PHI both going up two spots.
Six teams would have dropped in the standings — DET, IND, CHA, NOP, TOR, CHI — with the biggest drop coming for CHI, the only team to drop two spots.
• The only change to the playoff teams would have been a swap of PHI (11-10) for NOP (9-12). Using W/L, PHI actually would go from seventeenth to fifteenth overall, leapfrogging CHA (10-11) and pushing them down to the last playoff seed. NOP would then be eliminated entirely from the playoffs.
Top five seeds, 9Cat: NYK, CHI, MIN, UTA, DETTop five seeds, W/L: NYK, MIN, UTA, CHI, GSW
The big change would have been GSW overtaking DET for a top five seed, other than that, it's just a bit of shuffling for seeds #2-4, while NYK remains #1 with a bullet either way.
• Of course, for W/L, there are many more tiebreakers to figure out, as there were ties between teams at 18-3, 17-4, 14-7, 13-7-1 and so on. That could be more of a headache to figure out, in theory, but since the first tiebreak is probably head-to-head between the teams in question, it could make those matchups even more important. (Actually according to the charter the tiebreaks are divisional record and then head-to-head?)
• In both 9Cat and W/L, we had fifteen teams that were above 0.500, so there’s no change there. Also, there is no change in the bottom seven seeds, as the worst to seventh-worst teams remain the same: MIL, PHX, SAS, MIA, OKC, SAC, ATL. Assuming this season as a standard, the race for the worst record in the league would be about two to five wins.
Pros for W/L Method:
1. Seeing a W/L record is much nicer than a huge 9Cat WLT number. For the purposes of articling and seeing win and loss streaks, it’s much easier to see a W/L record. “X team is on a five game winning streak!” This is honestly the biggest reason I’m moving for a change... I’m hand calculating W/L as I do articles and it’s not fun.
2. Strength of schedule is an issue. Since we only have twenty or so regular season weeks, it’s impossible to play everyone in the league once, meaning at most we are facing off against seventy-five percent of the other teams in the league. That other twenty-five percent (estimated) of a team’s schedule is totally variable. Running into a few tanking teams can pad the 9Cat WLT numbers quite easily.
Actually, our schedule is two games each against our division mates, which is eight games total. So for a twenty-one week season, that leaves thirteen games versus the other twenty-five teams that aren’t in your division. That translates to twelve teams that you won’t even see each season.
Tanking teams basically give up a free 1-8 or 2-7 win, which inflates the 9Cat records of their opponents by quite a bit. There are too many tanking teams — not necessarily a negative, just saying — and with eight out of thirty teams in pure tank mode, plus a few more teams winning less than a third of their games, RDA is just about one-third pushovers and two-thirds other teams. For a playoff level team, every three weeks is a free 7-2 or 8-1 win practically. With that being the case, I think a win should just be a win, without considering 9Cat WLT.
Cons for W/L Method:
-1. An immediate concern for considering moving to W/L versus 9Cat is that since the league started out as 9Cat, it could be treacherous to move to W/L now. It would obviously be a huge change, but as I hope to prove, it’s not actually that big of a difference.
-2. The few teams that were potentially specializing in five-cat, multi-punt, strategies have mostly shifted away from that. While super big -- or super small -- ball is theoretically viable, I don’t think we have any actual teams building in those directions anymore. (In my eyes, this is the biggest issue with switching from 9Cat to W/L, the fear of an all big or all small team.) However, with our teams more or less assembled, I don’t think we should fear a pure five-cat strategy that would upend things. Or am I wrong and going toward W/L is a chance for someone to unbalance the whole league?
-3. It is pretty interesting to have a regular season that is played slightly differently than the postseason. There is some nice pressure to win as many categories as possible week-to-week under 9Cat. Were we to switch to a W/L system, the pressure could be off for teams to find as much balance as possible between the nine scoring categories. So the question is: Should the regular season and postseason be two different types of games? Debate!
Strength of Schedule (SoS)
Here is a chart of the teams that faced the easiest schedule, weighted by opponent category wins on the season:
It’s impossible to control for strength of schedule of course, but there should be some correlation between SoS and wins, as thirteen of the sixteen eventual playoff teams faced average to below-average SoS. The three playoff teams that didn’t were CLE, BKN, and NOP.
Note: Of course the bad teams tend to feed their opponents categories, thus pumping up their numbers, but I’m not sure how to correct for that. Running the numbers by W/L wins instead of 9Cat wins resulted in similar results, as you can see here. [ SoS Full Chart ]
The average number of playoff bound teams someone should face over a twenty-one week schedule was 11.2 playoff teams. However, ten teams had less than that — with seven teams facing ten playoff bound opponents and three facing nine or even eight (UTA, NYK, CHA).
On the other end of the spectrum, five teams got pummeled by future playoff teams. SAC faced off against thirteen future playoff teams while DAL, LAC, OKC, and MIL all had fourteen such opponents. Of course, it could be argued that these last few teams fueled their opponent’s playoff chances but I don't know how to correct for that either.
The five teams facing the toughest schedule this season comparing between systems:9Cat: MIL, OKC, TOR, BOS, PHIW/L: MIL, OKC, TOR, BOS, LAC
As you can see, there's not much difference, although it’s impressive that PHI faced such a rough schedule but was right on the bubble for the playoffs — even making it had we counted W/L instead of 9Cat WLT. Good job Joachim and Alvin!
Crushers vs Cream Puffs
And finally, let's look at some teams I've termed "crushers" versus "cream puffs." This mostly serves to illustrate how some wins are way too easy as some teams are simply layups and are inflating 9Cat win loss records. If you schedule contains a lot of cream puffs, prepare for a blowout win...
For example, in fifteen of their twenty-one weeks, MIL gave up a 1-8 or 2-7 loss. PHX and MIA weren’t far behind with fourteen and twelve total weeks worth of getting creamed. BOS and LAC round out the top five most-lopsided losers with ten 2-7 or worse losses apiece.
On the other hand, NYK led the way with twelve weeks of dominating wins — defined as 7-2 or better — followed by CHI and UTA with eleven crushing wins apiece, and then DET and LAL with ten each. Would they have won those weeks anyway? Of course, but is a win is a win or is a beatdown worth more?
• The only four teams to never get crushed in any week’s matchup? CHI, GSW, NYK, and MEM. Congrats to these teams who always put up a good fight.
• Two teams who never felt the satisfaction of a crushing win: MIL and SAS
• Two teams had 9-0 victories: NYK was the first to do it -- versus ORL in WK11 -- and then surprisingly, LAC pulled off the same feat five weeks later in WK16 versus PHX.
• Three teams suffered 0-9 or 0-8-1 losses: LAL to UTA in WK7 (the 0-8-1) and then ORL and PHX to match the winners above.
