The Solution Part 2
Part one of the overview of The Solution dealt with the creatures. This will deal with the spells and lands.
Repulse: While there's no question that all four copies of Repulse stay in the deck, I suspect many people are too reluctant to sideboard it out. Don't forget that Repulse is just a cantrip bounce spell. In many matchups, there will be far more good creatures than there are slots available for them, and this is one good way to get a slot. While Repulse can be used to save a creature and draw a card a lot of the time, it often costs time to do that. In addition, there are a lot of people who will name Repulse with Meddling Mage on the theory that Repulse is the only card that can remove the Mage, so anything else named will simply cause the Mage to get bounced if they guess correctly. It's not a bad theory, but it's definitely worth the risk of taking out Repulse in that case. The problem comes when the opponent knows for sure that this is happening, at which point they shouldn't come out... and so on. This is only something to worry about if this move becomes common practice. In other matchups, the same problem occurs. Having Repulse in the deck against control decks that use only Nightscape Familiars is far inferior to simply having additional creatures or cards like Pure Reflection. Using Repulse defensively in essentially offensive matchups is a road to disaster, and Repulse is mostly defensive for this deck. Using it defensively in defensive matchups is fine. In mirror matchups in particular, keep in mind that it will be a long-term battle of permanents and this doesn't do enough to change that. Repulse is also (of course) a natural combination with Meddling Mage, but the Mage is almost always busy naming removal.
Exclude: This is also a very nice card that can be difficult to take advantage of. Clearly it clashes with Lashknife Barrier, for example. The problem is that to use Exclude properly, the deck needs to hold back the mana on turns when it wants to be casting creatures instead, with the other issue being that there are decks where Exclude does nothing or practically nothing. If nothing else it can be used on creatures that are about to be countered, especially by an opposing Exclude, but that's extremely poor. Drawing multiple Excludes can virtually force the deck to hang back to use them, which can be extremely risky. The simplest issue is if the opponent decides to cast a removal spell or a Blurred Mongoose instead, which puts the deck in the same position next turn, but even further behind. The basic conclusion is that while Exclude is key to the success of the deck it shouldn't depend on it when it doesn't have to. It's better to plan on casting spells for a while and then sitting back on Exclude than to sit back on Exclude first, unless there's clearly enough land and time to play it the other way. In the matchups where the power of an Exclude is needed the most, it doesn't do its job. Between all these problems, I decided to run only two Excludes in the maindeck. The other two were sideboard cards to take out red-green and other random creature decks. Against bad players, Exclude is excellent. Exclude should come out of the deck any time it can't be used consistently, since there are always better options waiting to replace it. When there is a question of whether it stays or goes, it's best to split the difference unless Exclude is going to be named with Meddling Mage. Before, Exclude fit perfectly into the deck because of Fact of Fiction and there were few alternatives. Now, it is a card that stays because it is too good not to use despite the alternatives. A final note is that Exclude should not be held back early on, even if the creature doesn't seem to be significant. Only a Lashknife Barrier already in play is a good enough reason to let something slide into play unless the mana for the Exclude looks like it is going to be available for the rest of the game without incident.
Absorb: This never gets touched, for obvious reasons. I've actually started to want to make my opponent lose three life more than I've wanted to gain three life recently, which is the reverse of the situation before. Still, there's nowhere near enough black mana to consider Undermine (see Dromar's Charm). Absorb doesn't generally need to be saved too carefully except against heavy control decks, where its job is to stop the Void or similar spell that gets the opponent back into the game. It gives at least a full turn there and often much more, and should be saved. Just about anywhere else, they can be burned whenever necessary or convenient, and if the life points seem necessary it often pays to just make sure it gets cast on something reasonable while the mana is available. This is especially true when at low enough life that burn in response starts to become a concern. I would consider cutting this for any reason positively insane.
Dromar's Charm: No one's going to argue that Dromar's Charm isn't a great card, but it requires black mana. Most versions will use one Salt Marsh and four Caves of Koilos, which gives five sources, and the first Dromar's Cavern is relatively painless. That gets the deck up to six. Realistically the most the deck can have is about ten, and that's going to hurt. I'm not comfortable with that. Even if I was, it would only be worth it for a lot more than just the Charm, and that's a lot more to be comfortable with.
Fact or Fiction: Here it comes. Don't play Fact of Fiction. The origin of this was in the testing for Day 3 of Pro Tour-Tokyo. Ben Ronaldson came up to me, and said "This is going to sound crazy but..." and proceeded to suggest that I should sideboard out two copies of Fact or Fiction in my quarterfinal match against Federico Bastos. Of course I did think for a second that he'd gone crazy, but he explained it and it started to make sense. If a Solution deck gets into a long term duel of firepower with a red-blue-black deck such as Bastos's deck, it will lose that battle every time. The way I won was by playing threats that he couldn't answer, so every creature I could get was better than trying to trade Fact of Fictions with a deck better prepared for it. Instead, I squeezed every creature I had into the deck, and won the match easily. The same holds true now. Against a control deck, trading end of turn Fact of Fictions will lose the game. As Darwin Kastle put it when we talked at the airport returning from Denver: "Can your deck go 'land, go' as well as my deck can go 'land, go'"? The answer is that it most certainly cannot, and therefore it shouldn't try. On the flip side of the coin are the matchups where The Solution plays control. Here the opposite happens. The way The Solution loses these matchups when it isn't mana screwed is when it chokes on Fact or Fictions and therefore can't keep up on tempo. It's true that without the Fact of Fictions, there's a danger of the deck running out of cards to play, but making the creatures in the deck more effective works fine as a substitute and therefore Lashknife Barrier can take the place of Fact or Fiction in these situations.
That doesn't mean that there aren't situations where this hurts. In particular, it can become problematic when the identity of the beatdown player is unclear. Those are the situations where there's a need to reload. Despite that, after sideboarding Fact of Fiction is less important to me than cards like Gainsay or Disrupt. Doing the math on space, that means that even with extreme sacrifices, Fact of Fiction doesn't get into the deck. I didn't even run all four Lashknifes after sideboarding in mirror matchups, where it's clearly a very good card. The only real situation comes against the worst matchups for the deck, like David Price's Denver deck. Here again, there's the issue that you can't really win a long-term card fight against Prophetic Bolts and Flametonge Kavus with just Fact of Fictions and therefore you shouldn't try. Gainsay going into both decks slows both sides down, which is another reason why going into control mode is a mistake and why I had four Disrupts and only three copies of Gainsay. If there's a long-term fight here it would have to be based on Crimson Acolyte to neutralize a few cards and followed up by other problematic permanents. But clearing room for the Acolytes would be tough and not having them removed at the wrong time would also be difficult. Still, this is the best argument for their inclusion if nothing better can be found; I have never tested this matchup. The idea is still clear, the fight must be based on permanents and/or tempo strategies or it will fail. In addition, the extra lands from Fact of Fiction just weren't making that much difference, since the deck can function fine without more than about five lands. It's useful to have more, but it doesn't make that much difference in most situations. In some special ones it does though. So if mana often doesn't matter, the Fact of Fiction will simply give the two worst of three spells, which isn't worth the time.
Disrupt: It's safe to put Disrupt in the maindeck, because every deck in the format plays enough spells to ensure that it can get used. In a desperate case, it can be cycled off of a Repulse, Exclude or Absorb, but I've never had that be necessary. The one time I had it against red-green it countered a Scorching Lava on a Spectral Lynx, which helped a lot. It also fits perfectly into the deck's mana curve plan, since it protects a creature on turn three when the mana would otherwise go to waste and the opponent is unlikely to have a spare point of mana. This is as opposed to Gainsay, which doesn't fit in the curve of the deck all that well. The problem with Disrupt is that its effectiveness declines once it becomes expected, and the effectiveness of not having one in hand definitely increases once Disrupt is expected. At this point, the deck should have at least two, three makes a lot of sense and four would be far from crazy. One thing about Disrupt is that while people will be willing to play around Disrupt when they can, almost no one will be able to afford to play around two of them. Most of the time it won't even occur to them to try. Similarly, once the first Disrupt is 'cycled' on some random spell, the second one will come as a surprise. I used that trick all the time: Use one of the two Disrupts, and they'll play into the second one. The problem with fitting in four copies of Disrupt is that Disrupt is a lower power card than what it will be replacing, and often only cycles in the late game. What that means is that Disrupt is going to drag down the long-term power level of the deck, and will prove weaker in the late game. In general, Solution is being pulled in two conflicting directions: The deck can be seriously strengthened, but it leaves a long term weakness that causes the deck to run out of spells and choke on land. This problem always has to be watched for.
Lashknife Barrier: I keep talking about it, and I've actually explained it already in reference to the cards I'm taking out to fit the Barriers in. They're vital to the success of the deck, and allow the deck to pull off things it could do no other way. The one thing that remains is to answer one of the questions I get asked way too often, which is why I didn't use Angelic Shield. The short answer is: "Lashknife Barrier is a cantrip." That should be all the explanation that is needed. Its effect is also more powerful. For example, it takes only one extra Goblin token to take out a creature under a Shield, but those tokens do nothing in combat against a Barrier. In general the danger of multiple damage sources outweighs the problem of Dromar's Charm still removing threats and Planar Despair still wiping out the board, which it would probably do anyway. What's left is then a comparison of casting costs and the ability sacrifice Angelic Shield. Sacrificing it can be useful, but it erases the previous effect. It's like playing a non-cantrip bounce spell. Drawing a card is much more powerful. In addition, the casting cost of the Barrier in many ways is better for the deck than the one on the Shield. There's plenty to do on turn two, and the deck is too dependant on colored mana as it is. I think this decision is completely clear. To top it all off, when sideboarding the Barrier or Shield comes out in those matchups where the Shield is likely to get sacrificed on anything but the last turn.
Finally, a brief word on the mana. It seems to be generally accepted that the deck runs a little more white than blue to compensate for the fact that four plains became Caves of Koilos. The question is whether to play an additional dual land or tri-land or two. Dromar's Cavern and Salt Marsh are the options. I decided that I was willing to play a Cavern more than I was willing to play a Salt Marsh. The deck already has a one drop and four Coastal Towers, and I'm loathe to use another tap land. On the other hand, the first Dromar's Cavern is mostly harmless now that Lashknife Barrier means that the deck doesn't need to hold back for counters on turn three. The two changes are not independent; they basically follow from each other.