Power Level Data
Commander (EDH) forum
Posted on Oct. 23, 2021, 11:21 a.m. by adboyer
An interesting thought experiment.
Hypothetically, if their power scale is correct then the average should come out to around 5 since their rating system has 5 being what they consider par, typical, or average. Based on the data set, we can be 90% certain that the average power level of the sampled decks lies between 5.19 and 5.20. Overall, this means that the Nitpicking Nerds method of rating commanders tends to slightly overestimate the power of any given deck, although some would say the difference is marginal.
October 23, 2021 11:24 a.m.
Let me know what you folks think of this little experiment. Here is a link to the summarized data https://tinyurl.com/RULE781 and if you have any questions about my methodology feel free to ask.
October 23, 2021 11:26 a.m.
RNR_Gaming says... #5
Edhrec is not a good metric for power levels. The site more so focuses on popularity of commanders and cards that go with them. While Atraxa, Praetors' Voice maybe among the top 5 most popular commanders of all time according to the site she'll always hover around mid-high power - it provides no immediate value other than a good body with some fancy keywords, doesn't do much of anything unless you're already ahead and other commanders in that same color combination are much more powerful.
October 23, 2021 1:22 p.m.
Grubbernaut says... #6
He didn't test the most popular commanders, though. He tested any with 100+ decks, and while you can say that's arbitrary, I think it's a fine point of reference. If nobody is playing a commander at all, even casually, it's almost certainly unplayable.
Theoretically, that'd also probably make up for the slight point difference, most likely.
October 23, 2021 3:37 p.m.
You are correct in saying that EDHREC is not a website designed for measuring power. What I did is use the number of decks that any commander had and then tallied those under ranking to that commander that the Nitpicking Nerds gave it.
For instance, Atraxa at the time had 6381 decks and was rated as being a power level 9 commander. Hypothetically this means that there are 6381 decks that have an average power level of 9.
By doing this for a large selection of commanders I was hoping to see the distribution of power rankings across EDHREC by using the commander power rankings given by the Nitpicking Nerds.
October 23, 2021 3:39 p.m.
Also, Grubbernaut, that might be something I could tinker with is next time including commanders with less than 100 decks to see if that brings down the average to closer to 5. I hadn't thought about that, so thank you for the comment. I originally restricted the data that way so data entry would be less time-consuming.
October 23, 2021 3:51 p.m.
RNR_Gaming says... #9
The best metric would be getting tournment results but as it stands edh is a casual format and there would be a good chunk of rouge/jank decks mixed in. There's also the multiplayer variable - even if someone is playing the best deck at the table they could get over policed by the table and lose to getting 3 v 1ed. Additionally, tournment results and solving the format in a widely accessible manner would be unhealthy for the format.
October 23, 2021 5:16 p.m.
RNR_Gaming, I 100% agree. My hope was that the ranking system used in this video https://tinyurl.com/2pzvu4n7 would be a decent enough approximation. Do you think that the way they rank commanders is accurate enough for a rule zero conversation for instance?
October 23, 2021 5:28 p.m.
RNR_Gaming says... #11
If you apply the same criteria to say a planeswalker or a really awesome uncommon (some do this in pauper edh) it should be roughly the same; I'm assuming that's what you mean by a rule zero conversion/or are you just referring to talking about your deck before a game?
They did bring up a point about fun. Fun and power are not synonymous. I don't enjoy playing storm but I cannot deny its' power or inevitability - I'd just leave that out of any accessment.
October 23, 2021 5:55 p.m.
RNR_Gaming I am mostly referring to having a discussion about my deck before a game to make sure everyone is on the same page. Basically, the power rating they assign to a commander is my jumping-off point for evaluating my deck's power. For instance, they rated Brago as a power rating 5 commander but since my Brago deck is a slower token spawning deck with no instant win combo's I would probably rate my deck as having a power rating of 4. As for how they evaluate fun, that's entirely subjective so I consider it but I don't put a lot of credence to it.
October 24, 2021 11:48 a.m.
RNR_Gaming says... #13
Oof Brago is one of the most inherently strong commanders. Even if you dull it down he can provide so much value. Granted, not as good if you're not playing stax as he's excellent for breaking parity but the ceiling on Brago is much higher than Atraxa. Brago is also a commander that usually gets hated off the table.
October 24, 2021 1:29 p.m.
RNR_Gaming says... #14
Basically, when you sit down at a random table most will assume the ceiling. So, even if you explain your deck, there may be a circumstance like a nut draw where you're able to turn 2 Brago and start cracking out value as soon as turn 3. I'd actually say Brago starts at a solid 7 and would need to be significantly toned down to be casual; but that's subjective still.
October 24, 2021 2:51 p.m.
adboyer 1st of all total respect and props for not blindly following the YouTube gods, you noticed a man behind the curtain, you had the intelligence to question, and the courage to post. These are rare qualities in modern times
Yes, you are correct their math is flawed
As RNR_Gaming pointed out your math is flawed becouse your source is based on popularity
Harvesting tournament data would be more accurate but still flawed becouse it's just a filter that started from popularity
Truth is all the math is flawed
commander power level is no different then any card power level meaning it does not accurately reflect the powerlevel of the deck as a hole and there for should not be the focus
The real math for a cards powerlevel is on a per deck & per meta basis so to figure it out
PER DECK
Worse Interaction:
How well does this card perform alone
Best interaction:
How well does it perform with support of another card
Probabilities of interaction
how many cards in the deck interact with it
PER META
Best timing:
how well does the card perform if played a situation needs it
Worse timing"
how well does the card perform if played when no situation calls for it
Demand:
based on the people you play with how often will situations need this card to be played
Good luck!
October 25, 2021 8:09 a.m.
Thank you RambIe, RNR_Gaming, and Grubbernaut for the constructive comments. I now realize that I made too many logic leaps by comparing apples to oranges, or in this case assuming popularity is a predictor for power. Additionally, since the power level of a deck is determined by more than the power level of the commander I made another incorrect assumption.
However, if we assume the power rating system I was attempting to test is valid, which may or may not be true, it does show that commander popularity isn't directly correlated with commander power. If nothing else, perhaps this shows that the format isn't skewed towards building around hypercompetitive commanders or super casual commanders. This could potentially indicate a high diversity of decks, which I am very glad to see.
October 25, 2021 1:57 p.m.
Grubbernaut says... #17
There's never going to be "perfect" data for something fluid and dependent on pilot skill, but there's nothing inherently wrong with the method you used.
It's a good indicator of what a typically-played list from a given commander will rank in relative power level, which should be a good frame of reference overall. If you wanted to get more picky, you could try to find a way to compare cEDH decks, but that's a can of worms.
In short: No need to apologize, this is a good and interesting data point.
October 25, 2021 3:57 p.m.
RNR_Gaming says... #18
It's an interesting topic but you're going to pull your hair out trying to come to a conclusion that makes the majority of most people happy. There are a lot of variables and conflicting opinions and most of the time it's just better to play the game - I've found numbers are often are not a good way to indicate power level; explaining what the deck does and what it contains is usually the best option for the rule 0 talk.
October 25, 2021 4:10 p.m.
adboyer don't stop!
learn, adapt, keep going. o and add more sources
edit: even if it isn't perfect you might be onto something more accurate then many are currently using
October 25, 2021 5:07 p.m. Edited.
KayneMarco says... #20
One other thing that should be noted about commander power level is the player variable. One could be using the highest rated cedh deck out there but if the player doesn’t know how to pilot the deck then that deck will lose almost every time to a less powerful deck. I’ve seen someone dump over $5k into a deck that was netdecked and then be all pissy because they’re losing more than they win all because they don’t know how to pilot the deck. Player skill level is key.
October 27, 2021 10:12 a.m.
KayneMarco, you are very correct on that. I have been on both sides of that situation before as a new player and now as a more experienced player.
October 27, 2021 10:44 a.m.
I have come up with another idea. I made a new ranking system that takes the average of two already established ranking systems. The two systems that I averaged can be found using the following links https://tinyurl.com/e2ydj3, http://bit.ly/RankingEveryCommanderSheet
Here are the results in a spreadsheet for all commanders on EDHREC with 100 or more decks to their name https://tinyurl.com/63tvk5ze
Let me know if you folks agree with the ratings for commanders that you have experience piloting.
adboyer says... #2
Hey folks, I ran an interesting thought experiment and I thought I would share the results with you all. I came across the videos posted by the Nitpicking Nerds where they rank commanders, and I wanted to see if I could use data to support or refute their power rating system. To test this, I took a snapshot of the EDHREC data of all the decks whose commander had at least 100 decks to their name.
October 23, 2021 11:23 a.m.