Nobody has shown yet that Thoughtseize taking your best card in a subpar draw creates any more non-games or negative variance than Bolt killing your T1 Birds in a mana-light draw or Mana Leak countering your one threat that you spent your whole turn to play. Those are effects that AFAIK everybody here not only doesn't object to but actively tries to pack into their Cubes. Thoughtseize suddenly being the symbol of everything that's wrong with power-max Cubes is baffling to me. The utility of discard in particular scales with the format's power band - if most cards are 3/10s but some are 8/10s, sniping those is crucial - and it's more important in a high-powered environment where cards' inherent power let them dominate games by themselves, but that doesn't mean that Thoughtseize itself is oppressive "in the same way" at all. Opposition is a platinum hit in any format you put it in. Thoughtseize is... a good card that you'll take in the middle of the pack if you're in black. That's true in Holiday Cube and it's true in Theros Limited.
This analysis of "bolting the bird" also spoiling games completely misses the fact that your bolt does nothing to my
Wall of Omens,
Wrath of God,
Farseek,
Mind Stone, or any of a plethora of cards I could have in hand that are critical to me being able to do anything after a shaky mulligan. It also ignores the fact that keeping a 1-lander and a
Birds of Paradise is a very risky hand to keep, and that keeping that hand is a calculated risk. I can keep
Birds of Paradise on a one-lander and make that calculated risk, knowing full well that a T1
Lightning Bolt will undo me. But those aren't the only sorts of games that
Thoughtseize will ruin, and
Thoughtseize ruins the "bolt the bird" game in a way that Bolt doesn't, by forcing me to reveal my hand.
Let's move beyond the bird roast and look at other times that
Lightning Bolt won't replicate
Thoughtseize in any way. I can also keep, for example, that 2-lander and a
Wall of Omens or a
Farseek and hope for the best, or 4 lands, a
Thraben Inspector, and a
Wrath of God and feel comfortable enough in my do-nothing hand that I probably won't lose before drawing into something else because, hey, at least I have this boardwipe if I start flooding.
Thoughtseize decimates each of these perfectly reasonable keeps in a way that
Lightning Bolt doesn't, because 1. it can strip any non-land card, whereas
Lightning Bolt has much finer parameters for what it can answer, and 2.
Thoughtseize insists being played out immediately for maximum value, whereas
Lightning Bolt is significantly more flexible about its timing, and 3.
Thoughtseize not only answers one of my cards in-hand,
it also lets you know what else I could possibly play, allowing you to craft some great follow-up turns based off of this information and denying me the chance to bluff my way forward with a bad hand.
Sure,
Mana Leak can similarly wreck most of these hands; but that's matching
my cards with
your answers in an appropriate fashion, because you paid the opportunity cost of leaving mana up to counter, and you take a calculated risk Leaking my, say,
Farseek, not knowing that I'm land-light and counting on
Farseek to keep me in the game.
Thoughtseize not only
denies me all hope in response to each of these hypothetical, reasonable mulligans, it also
lets you know how screwed I am.
Mana Leak lacks this feature. I don't know about you, but if I'm on control, I'm not
Mana Leaking a
Wall of Omens typically; I'm saving my counterspells for something a little spicier. And ignoring that play means my 2-land hand digs deeper into my deck for land, which might just allow me to recover from the less-than-perfect mulligan I kept. And these aren't pie-in-the-sky, you-were-never-gonna-win mulligan keeps, either! These were perfectly acceptable hands, that were wrecked before the game even had a chance to get underway.
Thoughtseize in each of these scenarios is such a significant blow, that even comparing it to a
Counterspell/
Murder hybrid doesn't do it justice, because you
also get to see how bad my hand is post-removal. I really can't stress this enough; it gives you a full license to play out aggressively as you please and run me over because you can make certain assumptions or calculations about where my hand
might develop, and it's all thanks to this one little card. It's a very powerful effect.
(sidenote: I hate these Amonkhet invocation arts appearing on mouse-over now... I wish we could get rid of them!)