Decks to Beat (2020-21) Discussion Thread

Anyone interested in this any more?

I wouldn't mind upkeeping it. I don't have access to the actual thread, nor can I move decks though.

Also, I am not so sure about the cut-off. As best I can tell, it was arbitrarily chooses to be 3.4%. Might be good, might not, hard to say, exactly, but I'd wonder why not just a 3.5% either? But then again, we can play that game all day until 4 or 5%.

In any case, here is the January 2020 data:
https://i.imgur.com/Lh4Jg6w.png

https://i.imgur.com/dbRoGBD.png

Which means 3.4% gives a cut off of (rounded) 108 points. I'd be apt to consider 3.5%, which is (rounded) 111 points.

I'd say TES gets in, Dark Depths just misses.

Top decks would be then:
UR Delver
Miracles
Death and Taxes
Eldrazi
Grixis (Delver)
4C Control
Hogaak
TES

Thoughts? Obviously I need to wait until February is well and over to compile those results.

I appreciate the effort. Would be great if this was kept up to date. Oko doesn't seem to be much of an issue in legacy, just as I expected.

I'd like to do whatever I can to support this.
Even if it means passing the torch off.

I think I remember that originally, it was done so there would be 5-6 decks to beat at a time. The metrics were all kind of wish washy before, so maybe a deeper look into that might be warranted.

Quote:

---

Originally Posted by UseLess

I appreciate the effort. Would be great if this was kept up to date. Oko doesn't seem to be much of an issue in legacy, just as I expected.

---

Well, I think it and Veil are doing something, but are they dominating? No, not really. Keep in mind that most of the Miracles lists are running Oko themselves.

But, it hardly seems that Oko, Veil or Astrolabe are dominant. In fact, it seems the format has adjusted to either playing them, or fighting them, fairly efficiently.

Thanks for doing this H! It'd be great to have the DTB section up and running again.

Quote:

---

Originally Posted by UseLess

I appreciate the effort. Would be great if this was kept up to date. Oko doesn't seem to be much of an issue in legacy, just as I expected.

---

I would guess most of those "Miracles" decks are Bant Miracles with Oko.

If you look at the Feb results so far, according to MTGTop8 and MTGGoldfish, the top 3 decks are:
UR Delver
Breach
Snowko

Both Bant Snowko and 4C Snowko are top decks. Astrolabe Oko decks are doing well but not running over the format.

It certainly seems like the following might be happening though:

Oko is being played, which means Red Blasts are really good. The decks taking advantage of that are UR and Grixis Delver.

Since many decks are playing Astrolabe and Basics, Wastelands are somewhat low, so 12 Post/Eldrazi is doing well, not to mention Moons are also low, so there is that lack of threat too.

Veil's effect is especially unclear in the data, but it might be the case that it is helping prop up Miracles lists that would otherwise we weaker to combo and/or discard?

Yeah I realized after making my post that the miracles lists probably contained some amount of Oko's. Still, not a problem card, just a strong one in the right decks. That meta list looks quite healthy to me.

To me Veil shows up the most in seeing ANT and TES having swapped dominance in the results.
For a while ANT was DTB or just barely out, with TES kinda middling behind somewhere in the masses.

Now that both your win condition in Tendrils is suspect, along with your main axis to remove the card in the form of discard is suspect, ANT has taken a hard hit.
Meanwhile the less consistent poop out 16 Goblins deck is safer as a result. Also I've seen a number of Echo of Eons type decks being lumped in as TES, not sure what that does to the math.

But that's the most telling sign that something is up in the cardpool to see ANT drop so hard without taking all of storm decks down with it.

Quote:

---

Originally Posted by PirateKing

To me Veil shows up the most in seeing ANT and TES having swapped dominance in the results.
For a while ANT was DTB or just barely out, with TES kinda middling behind somewhere in the masses.

Now that both your win condition in Tendrils is suspect, along with your main axis to remove the card in the form of discard is suspect, ANT has taken a hard hit.
Meanwhile the less consistent poop out 16 Goblins deck is safer as a result. Also I've seen a number of Echo of Eons type decks being lumped in as TES, not sure what that does to the math.

But that's the most telling sign that something is up in the cardpool to see ANT drop so hard without taking all of storm decks down with it.

---

TCDecks categorization does indeed leave much to be desired, honestly. Most of there aren't too bad, but some of them are kind of just lumped together, or named strangely.

Thanks for posting! I utilize the Source quite a bit as my primary outlet, along with the LaL facebook page and the local yokels.

Quote:

---

Originally Posted by PirateKing

To me Veil shows up the most in seeing ANT and TES having swapped dominance in the results.

Also I've seen a number of Echo of Eons type decks being lumped in as TES, not sure what that does to the math.

---

Not much. About 50% of those TES results are Bryant Cook.

His new build is on a hot streak. He made some big changes from older TES builds to adapt to the meta (Wishclaw Talisman over Infernal Tutor, Mox Opal, Defense Grid, Veil, Echo), but by definition whatever he's on is "TES". It still uses the signature cards of Rite of Flame, Burning Wish, and Empty. TES adapted to the meta. ANT hasn't.

That also makes the results hard to interpret. Does it mean TES is that much more dominant in this meta, or does it just mean Bryant Cook is good at storm?

Quote:

---

Originally Posted by FTW

Not much. About 50% of those TES results are Bryant Cook.

His new build is on a hot streak. He made some big changes from older TES builds to adapt to the meta (Wishclaw Talisman over Infernal Tutor, Mox Opal, Defense Grid, Veil, Echo), but by definition whatever he's on is "TES". It still uses the signature cards of Rite of Flame, Burning Wish, and Empty. TES adapted to the meta. ANT hasn't.

That also makes the results hard to interpret. Does it mean TES is that much more dominant in this meta, or does it just mean Bryant Cook is good at storm?

---

No I mean, I've watched his streams, I understand the changes and why. Of course he'll be the best example of the deck. But even then, one man can't put up enough numbers to affect the results like that. It would take several players having similar success, and if we assume they are all not as skilled as their patron, then the deck must have something going for it.

But even beyond all that, I feel like you just said the same thing I said:
TES adapted to the Veil of Summer meta. ANT hasn't.
To which, I agree.

Quote:

---

Originally Posted by PirateKing

But even then, one man can't put up enough numbers to affect the results like that. It would take several players having similar success,

---

No, what I'm saying is it is one man.

Check out the MTG Goldfish Legacy metagame. TES is one of the top decks in the current meta.
https://www.mtggoldfish.com/archetyp...s-137062#paper

Then look to the right.
13/25 of the decks are Bryant Cook.
Most of the others are Vivarus.
It's just 2 TES pros wrecking the meta.

On MTGTop8, which includes paper events, it's harder to parse. There's a lot of "Storm", but most aren't TES. Even some of the ones called "TES" are really ANT (Cabal Rit, Infernal Tutor, PiF, no Rite).
https://www.mtgtop8.com/archetype?a=35&meta=39&f=LE
Of the 8 true TES decks there, 3 are Bryant Cook. They're missing his 3rd place @ Leaving a Legacy VI, making it 4/9.

He's singlehandedly half of the recent TES results.

Getting better source material than TCDecks provides would be good. So would transparency regarding what constitutes a "deck to beat."

I miss seeing the mods' take on what's good and what isn't, but I'm not pleased at the amount of sludge it fueled—and fueled with data ripe for cherry-picking.

Very nice!

Quote:

---

Originally Posted by H

Also, I am not so sure about the cut-off. As best I can tell, it was arbitrarily chooses to be 3.4%. Might be good, might not, hard to say, exactly, but I'd wonder why not just a 3.5% either? But then again, we can play that game all day until 4 or 5%.

---

I assume it was chosen for a historic reason, then kept to enable comparison between different periods.

Quote:

---

Originally Posted by H

Anyone interested in this any more?

I wouldn't mind upkeeping it. I don't have access to the actual thread, nor can I move decks though.

Also, I am not so sure about the cut-off. As best I can tell, it was arbitrarily chooses to be 3.4%. Might be good, might not, hard to say, exactly, but I'd wonder why not just a 3.5% either? But then again, we can play that game all day until 4 or 5%.

In any case, here is the January 2020 data:
https://i.imgur.com/Lh4Jg6w.png

https://i.imgur.com/dbRoGBD.png

Which means 3.4% gives a cut off of (rounded) 108 points. I'd be apt to consider 3.5%, which is (rounded) 111 points.

I'd say TES gets in, Dark Depths just misses.

Top decks would be then:
UR Delver
Miracles
Death and Taxes
Eldrazi
Grixis (Delver)
4C Control
Hogaak
TES

Thoughts? Obviously I need to wait until February is well and over to compile those results.

---

I sent a few PM to dice_box in order to have it updated... never had a response.
I'd be glad if anyone would do this again.

The bigger question is whether or not Decks to Beat would be an actual measure of anything.

Quote:

---

Originally Posted by Wrath of Pie

The bigger question is whether or not Decks to Beat would be an actual measure of anything.

---

That really depends on what we consider to be "actual" there.

I think it does serve a purpose, that purpose being to frame what seem to be the top performing deck of the previous month. That doesn't mean they necessarily are, but rather, simply we can capture that seeming.

In a sense, it gives us a notion of a meta to consider. From there, of course, we need to actually think further about what the expected meta would be for a given event. But at least the frame is notionally set. Like all frames, it fails to capture the whole of the picture, but it at least provides a structure to evaluate to and from.

Quote:

---

Originally Posted by FTW

No, what I'm saying is it is one man.

---

Not going to lie, that is so awesome :eek:

Well, to be 100% clear, I am not exactly a huge fan of the methodology I followed, but I also don't hate it, per se. I simply followed the previous formulation. I don't think TCDeck's parsing of the data is at all ideal in any sense, but what would be the other options?

If there was another way to aggregate the data, I'd be all for it, but I really don't want to be doing the whole thing manually.