Why do some people run fetches in their monocolored decks?
Commander (EDH) forum
Posted on Nov. 3, 2019, 1:58 a.m. by dbpunk
So I've seen this a few times, especially with competitive builds, that some people will run fecth lands in monocolored decks. Is there a reason for this? Like I can understand if you're running a fetch in a deck with more than one color, but i feel like it would be better just to run a basic or monocolored land with an ability at that point.
multimedia says... #3
Depending on the cards you play in your mono colored deck, Fetch lands can add a lot of utility. They add shuffling effects and here's some examples:
Cards that can put cards you don't want on top of your library ( Brainstorm ) or see/play the top card of your library ( Oracle of Mul Daya ). Cards that let you arrange the top cards of your library ( Sensei's Divining Top ). Having a shuffle effect can be very beneficial in seeing and arranging the top cards of your library.
Many cards after you select a card you put the rest that you could of chosen from on the bottom of your library. Having a shuffle effect makes it so those cards don't stay on the bottom of your library. Some cards let you play lands from your graveyard ( Crucible of Worlds ) and a Fetch land can give you a land drop via your graveyard as well as way to get a land out of your library each turn.
Of course there's many other ways to get a shuffle effect ( Demonic Tutor ), but getting one from a land which then gets you another land out of library is good.
November 3, 2019 5:26 a.m.
shuffle effects, feeding the graveyard, extra landfall triggers, the list is actually pretty long. Hell it's also pretty nice to figure out what the opponent took out of your deck with their pyrexian tutor or cards of yours exiled face down.
November 3, 2019 7:39 a.m.
Huh I never thought of it that way. Tbh that's really something I wouldn't have thought of (and yeah I now realize that doing the math it could actually be a significant change to statistically get the card).
I don't think I'll ever be that competitive, but interesting.
November 3, 2019 7:41 a.m.
I personally enjoy the synergy between fetch lands and Rings of Brighthearth
November 3, 2019 8:15 a.m.
MagicalHacker says... #7
They erroneously believe that they are more likely to win and less likely to lose if they have a 0.7% smaller chance of drawing a land than the chance they would lose or not win for having 1 life less. If you're playing commander where no one wins with combat damage, then that's the right move, but for us playing normal commander, it's a dumb move.
November 3, 2019 9:23 a.m.
Gidgetimer says... #8
Deck thinning (statistically improving your chances of drawing a non-land card) is a fallacy. All of the other listed reasons are valid. Landfall triggers, shuffling your deck, and graveyard interactions are all valid; but the statistics of drawing fetches and then that removed land drawing you an extra non-land card are miniscule.
Even in 60 card with 20 lands in a 12/8 distribution of Fetches to "true" lands the first statistical card isn't until turn 21. This is the same as running 18 fetches out of 30 lands in EDH. In the more reasonable 4/16 case (6 fetches in EDH) The first card isn't realized until turn 36. These numbers come from a Monte Carlo simulation where a "turn" is a card draw and a land drop, so you can make "turn N" into "N+7 cards seen". That is still 28% of your deck drawn in the extreme case or 43% in the reasonable case for a single card of advantage.
November 3, 2019 9:29 a.m.
Thanks for that Gidgetimer - saves me having to look up that link.
I think these days most people know that deck thinning isn't actually a thing, and you hit yourself for damage with fetches, which also puts you on a faster clock.
November 3, 2019 9:55 a.m.
Generally the decks that are running fetches in mono-coloured decks are not lists where much in the way of combat happens (as in cEDH) by comparison to typical pub stomp, so the 1 life lost is about as much of a moot point as the tiny bonus to deck thinning. But the rest of the points still stand.
November 3, 2019 11:57 p.m.
dingusdingo says... #11
The deck thinning from using fetches IS a real thing, and has relevance on probability to draw more non-land cards. The difference is more than marginal, and the statistics of drawing a non-land versus land are changed by using the fetch. To call deck thinning a fallacy is incorrect.
I've covered this extensively here before, and I'm going to CnP my response because it is relevant.
https://tappedout.net/mtg-decks/yeva-draw-grow/discussion/?&p=17&page=17
You are in a game where you have 15 fetchlands on the field, and your deck is exactly 100 cards (hypothetical!). You have 30 more lands in your deck. If you pop no fetchlands, the chance of drawing a land on any given draw is 30/100, or 30%. Now imagine you have popped all 15 fetchlands at the same time. You have grabbed 15 lands out while reducing the library size by 15, so your chance of drawing a land is 15/85, or 17.6%. If you don't want to draw anymore lands then the fetchlands have helped.
What if we only use 5 of the fetchlands? Chance to draw a land then becomes 25/95, or 26.3%. Our chances to draw a non-land are better with more lands removed from the deck, despite the deck size dropping overall.
To use an even more absurd example, imagine you have 3 cards in your library and one fetchland in play. 2 cards in your library are lands. If you use the fetchland, your chance of drawing the nonland goes from 1/3 (33%) to 1/2 (50%).
Combining our two examples, deck thinning increases your chance to draw any specific card in your library by reducing the overall library size. It is more effective with less cards in library, so every fetch that happens after another fetch or after any draw has greater impact on the % of lands to non-lands than any previous draw or fetch. You also have to consider that you're raising the chance to deck thin too. You have a higher chance to get a deck thinning land, and also greater significance from every deck thinning card added.
The downside to running the fetches are -1 life and possibility of getting caught by stacks pieces, giving them lower value than regular lands. In a 40 life format where people win by comboing, the -1 life is negligible in a deck that doesn't use life as a resource (i.e. Ad Nauseum). The chance to get caught by stax pieces is scarier, but the amount that specifically effect Prismatic Vista that WOULDN'T effect a basic Forest are:
- Thalia, Heretic Cathar
- Blood Moon
- Magus of the Moon
- Back to Basics (only if it enters tapped)
- Stifle
Also, many non-basic hate simply doesn't work on it due to timing. Ruination or anything that isn't a constant effect the Prismatic Vista can just sac in response.
You should 100% run Prismatic Vista in cEDH. And if God forbid they ever print a fetchland that fetches fetchlands, you should run that too.
99 cards in an EDH deck, start with 7 in hand for 92. You also get 1 draw a turn and 1 land a turn under normal circumstances. If we compare 3 turns of fetchlands versus 3 turns of regular lands I think it will make more sense. We will assume 30 lands, with 3 starting in hand, and that you don't draw any lands for ease of math purposes.
Regular lands
- Turn 1, after draw you have 27/91 for 29.6% land chance on any draw
- Turn 2, after draw you have 27/90 for 30% land chance on any draw
- Turn 3, after draw you have 27/89 for 30.3% land chance on any draw
Fetchlands
- Turn 1, after draw you have 27/91 for 29.6% land chance on draw. After cracking the fetchland, you have 26/90 for 28.8% land chance on any draw
- Turn 2, after draw you have 26/89 for 29.2% land chance on draw. After cracking the fetchland, you have 25/88 for 28.4% land chance on any draw
- Turn 3, after draw you have 25/87 for 28.7% land chance on draw. After cracking the fetchland, you have 24/86 for 27.6% land chance on any draw.
Analysis
Assuming no lands drawn and 3 lands in hand, the percentage chance to draw a land increases with every draw. While using fetchlands, the percentage decreases. After 3 turns of using non-fetches, there is a 30.3% chance to draw a land, versus the 27.6% for using entirely fetchlands. The difference after 3 turns ends up being 2.7%, which doesn't sound like a lot until you factor in performance across a large number of games. Having access to higher quality cards is just as important as having access to more cards.
The difference in % of card quality is interesting to look at too.
- Turn 1 there is a 0.8% difference.
- Turn 2 there is a 1.6% difference.
- Turn 3 there is a 2.7% difference.
The other thing to consider is how density of fetches is impacted by adding 1 more fetch. You currently run 4 fetches, with a 7.07% chance of starting with any given fetch in the opening seven. To put it another way, you have a 92.93% chance of NOT starting with any given fetch. We can calculate the odds of not starting with ANY fetches by using (0.9293)^x, where x is the number of fetches you run. With 4 fetches, you have a a 74.5% chance of not starting with a fetch in hand. If we bump that up to 5 fetches (adding Prismatic Vista , we instead have a 69.3% chance of not starting with a fetch.
Put another way, swapping a basic forest for Vista increases your odds of starting with a fetch from 25.5% to 30.7%, a 5.2% increase! You are increasing the chances you get a card that increases your draw quality, which is a great way of saying you're improving card quality.
5 Fetches scenario
Using the same scenario but starting with 5 non-fetches or 5 fetches instead. Assuming 30 cards of your 99 are lands, and no lands drawn during this period.
Regular Lands
- Turn 1, 25/92 for 27.2%
- Turn 2 25/91 for 27.4%
- Turn 3 25/90 for 27.8%
- Turn 4 25/89 for 28.1%
- Turn 5 25/88 for 28.4%
Fetches
- Turn 1 25/92 for 27.2% prefetch, 24/91 for 26.4% postfetch
- Turn 2 24/90 for 26.7% prefetch, 23/89 for 25.8% postfetch
- Turn 3 23/88 for 26.1% prefetch, 22/87 for 25.2% postfetch
- Turn 4 22/86 for 25.6% prefetch, 21/85 for 24.7% postfetch
- Turn 5 21/84 for 25% prefetch, 20/83 for 24.1% postfetch
Difference
- Turn 1 0.8%
- Turn 2 1.6%
- Turn 3 2.6%
- Turn 4 3.4%
- Turn 5 4.3%
Thoughts
Having a smaller chance to draw a land off the top deck is nice. It means you need less draws to reach the cards that make you win. Fetches also are incredible for color fixing purposes, which can't be overstated in a 3 4 or 5 color deck. Fetches are actually more useful in a 60 card environment too, because you can run higher density of color appropriate, as well as the smaller land count + card count means each land and card removed has a higher impact on your draw percentages.
One of the ideas that makes Magic really fun is how chance and probability play into it. In casual circles, chance helps to level the playing field between decks, as anyone can get hit by bad top decks. One of the things that separates casual and serious play is knowing how to build decks to navigate the % chance of chaos, as well as designing decks to work consistently despite chance being part of the game design. Most decks do this through using tutors (effectively copies 5-8 of your winning card), or by drawing many cards (increasing your chances to draw the card you need). Deck thinning is a staple of competitive decks in all formats because it gives you that small advantage which you then hedge against other decks. Even if the advantage is small, by repeatedly stacking every advantage in your favor, you are much more likely to win.
I wanted to make an addition to the above passages in the context of the link Gidgetimer posted.
The content you linked actually proves that deck thinning does have an effect. Look at the graph for the average number of spells drawn utilizing different amounts of fetches. Specifically, the 12/8 comparison shows an extra spell seen on turn 20 in comparison. I don't want to dive into upper/lower quartile and how the effect is bigger based upon usage in averages (games with no fetches seen factored in to their average performance), but even looking at just the average, 1 extra card seen per 20 turns is pretty significant. You may not see 20 turns in a single game, but you will see it across many games, which is where fetches are excellent. You're receiving a card quality advantage on a minor scale, but it is still an advantage. 1 extra card seen every 5 games or so still increases performance.
Of course the actual fetchland usage isn't 12/8 in EDH. It depends if you run Prismatic Vista and how many colors your deck is. Regardless, the point holds that you're going to gain a small but measurable advantage across a handful of games. Using the 4/16 appropriated to EDH, even one extra card across 36 turns is worthwhile. If your games only make it to turn 3, you're still seeing an extra card in 1 of every 12 games. Even at those numbers I would still run the fetch.
The author even notices the card advantage. The author just doesn't think its worth it in a 60 card format with 20 life.
"Ultimately, then, I would argue that the data bears out the contention that playing fetch-lands for their thinning effects are a bad idea: Only a suicidally reckless aggro deck can afford 4 life for a card, and those decks can’t afford to wait 20+ turns for it."
In the context of MOST games of Magic, the deck thinning is negligible. If you want to deck to perform its absolute best, then yes deck thinning will yield better draws across many plays. When you put the life loss into context, the disadvantage of the life loss can outweigh the marginal deckthinning benefit. This is what the author is saying. That doesn't erase the deckthinning from happening or impacting card quality.
As other people have pointed out though, its just one drop in the bucket. Fetches also cause a shuffle, trigger landfall, color fix, and form an engine with cards like Crucible of Worlds . Thinning is real and works, but you're most likely running fetches for color availability. The thinning is the cherry on top.
November 6, 2019 7:35 p.m.
Gidgetimer says... #12
I had a big long reply typed up, but at this point it just isn't worth it. There are enough articles out there that provide clear logical arguments for why there are no decks that thinning helps. If you want to ignore the shifting probability that necessitated the simulation instead of pure mathematical analysis there is nothing I can do. Just be aware that "1 card expected by turn 36" isn't the same as "1/36 chance to get an extra card on each turn". The probability of getting an "extra" card by turn 3 is closer to 1/200 than 1/12. And then there are considerations about if that "extra" card is actually going to do anything.
ZendikariWol says... #1
(well, only five of the fetches will do anything in a monocolor deck, but 10% is still pretty huge)
November 3, 2019 1:47 a.m.