The spell lands are proving to be rather better than expected. I put all of them into the cube for testing and none have come out yet, none even really look shaky as yet. They are not bombs but they add more to your deck than I imagined and are well worth a pick even when all they replace is a no-pick-required basic land. Cards like Teetering Peeks are great but they never last in cube because they simply don't merit the slot in terms of impact in games. Sure, you play it in your aggressive red deck and make it marginally better but such things do not lead to a better cube environment overall. The red player would always rather 24 good cards and 16 Mountains than 23 good cards plus a dodgy one, and a mana base including a land with minor upside. So what is making spell lands or these "modal dual faces cards" so much better than other utility lands and cycling cards?
While evaluating these cards I used cycling as my base line. Flip lands looked a fraction better than cycling for the most part and so I took Pelakka Cavern / Predation to be equivalent to Memory Leak in terms of power level. Leak is a slightly better spell but having a flipside tap land is better than cycling for one so it is roughly balancing out. The reality however seems to be that having a land on the back is rather more powerful relative to cycling than I had appreciated and I have been trying to work out what is behind this. Especially when cards that are good early but weaker late like Force Spikes and hand disruption are exactly the sort of thing that shine with cycling relative to the norm. You would expect the cycling to trump the spell lands in those cases but seemingly not thus far. It is certainly closer with those sorts of cards in that the difference between Censor and Jwari disruption is small while the difference between Ondu Inversion and Akroma's Vengeance is massive (I couldn't think of a closer comparison but it is close enough to make the point!)
The first and most obvious case is that cycling on a spell is only OK in a mana screw while a flippy spell land is great. When you need a land your Memory Leak only represents a 40% or so chance of being a land. Predation is 100% lock in. Both cost you a mana in that Caverns comes in tapped, the Leak might even cost two if it finds a CitP tapped land. It might even effectively cost you more if you didn't play something else relevant in the hope you would hit a land with the cycle and be able to do both but instead miss and do nothing.
This is also pretty much the second reason the spell lands are superior to cyclers - the certainty. When you cycle a card you are just tapping into the probability of what you might draw. You feel land light so you cycle so as to prepare for turns later down the line. That cycle might set you up well and it might not. It is also harder to know when it is right to be pulling the trigger on a cycler. The assured outcome of the spell land seems to make the choices a lot more clear cut and with a higher percentage of being correct. You can plan around the card in hand a lot more easily and formulate a long term plan. Cycling cards tend to sit in hand and muddy the waters in terms of planning. Indeed, they can effectively reduce your ability to plan precisely in the sense that you are one card down in hand that has a known outcome. Option density is usually a great thing but when it comes at the cost of certainty it is not all upside and that is a subtle hidden issue present in cycling cards.
The next difference is action density. This is certainly much more a thing in cube than other formats but it is absolutely relevant and is a big part of why I dislike cards like Tooth and Nail in cube. There is lots of draw effects in cube and games often go fairly deep into decks. You only have so many deck slots to spread across threats, lands, removal, etc. Every card you cycle away gives you one less of total piece of action. In a long control affair or good old midrange grind it can often come down to who actually has more total threats and answers in their deck. Cycling cards reduce this total while spell lands increase it when occupying a land slot. When occupying a spell slot the spell lands instead increase consistency. Either way it is a win both in general and when compared to cyclers.
The more I build and play with these cards however the more I appreciate that none of them really sit exactly in the spell slot or in the land slot and instead will be somewhere between the two based on the spell land in question and the deck it resides in. In much the same way that you can use the idea of cheap cantrip cards you can do so more directly with the spell lands (xerox theory). The generally accepted rate for cantrips is two to three cantrips at one or two mana equates to the ability to cut one land from your deck. This is a good guide although glosses over some subtleties, most of which are context dependent and command an article all to itself. There is slight diminishing returns on this effect for one, you cannot just gut a deck of land and ram it full of cantrips. You lower consistency after a point but before then you are really slowing yourself down, clogging up the works all the while reducing your action density significantly. Spell lands however let you you go pretty nuts on cutting lands if you wish. You can nearly get away with a one for one trade on lands to spell lands even with the ones that are better and more powerful spells. I have commonly found I am adding two or three to a deck and removing one less normal land overall for them. A normal deck might be 16 lands and 24 spells but with spell lands I will have effectively have a 17 land deck with 25 or 26 spells all without changing library size. This is obviously just great in and of itself. Even when we accept the spells on the spell lands are underpowered in general. Power is a function of mana cost which is less relevant than normal when your card can simply be a free mana source. As such you have two general groups of spell lands; the cheaper situational cards and the costlier higher impact ones. A 4/5 trample for 6 is laughably bad in terms of power level for cube but the point in the game where you are playing it means the mana cost is far less of a factor. You are still looking at a 4/5 which is still reasonably big and threatening. A lot of seemingly underpowered cards perform very well as spell lands just so long as they can have a big impact. The cheaper situational cards get their power from being very effective and thus cost efficient when the time is right or being something else if not in the more conventional way that modal and cycling cards can be powerful. It is a big win for this latter group of cards in cube that the present meta seems to have endless uses for mana. Clues, Castles, adventures, flashback, escape, kicker,.. the list goes on a long time. There is certainly diminishing returns on the power per mana spent on these various mana sinks but they are there in plentiful supply ensuring that your Jawri Isle is useful into the late game long beyond when a Force Spike is getting anything done.
What I have found is that having spell lands in my pool allows me to both play and build differently to normal. When your deck has 17 or 18 land you can expect to curve more consistently and play your 4 to 6 drop cards more reliably on or around their point on the curve. In essence the spell lands allow you to doubly increase your action density by having more powerful and more direct cards in your build. The consistency that I may have been getting from Elvish Visionary or Sleight of Hand is not as important as it was and so I can cut a card in that range and replace it with a planeswalker or something. Spell lands give you the possibility of the spells on the card directly but they also seem to allow for the cramming of more power over filler than before without harming consistency.
So why are these lands so much more exciting than the afore mentioned Teetering Peeks? Llanowar Reborn looks like it is double the card that Vastwood Fortification is being able to do both things rather than just the one. In the case of Fortification it is the power of the instant speed that makes most of the difference. When you can use it as a trick and blow out a combat or negate removal the value of the card jumps massively. When it is just making a dork a bit bigger it isn't worth the card. That is the same case for most of these kinds of utility lands. They simply don't offer enough of a return in terms of raw power and potential. The payoff is not only minor but generally also tied to when you make the land. It makes timing the land awkward and further reduces the value of the card. You have to play Teetering Peeks in a land slot and as a result it makes your mana base a touch worse. Your spell lands sit in a nice void between land and spell and generally allow for you to improve your mana base overall. I have no math to support it but I am sure 17 lands with a couple of EtB tapped lands curves out far better than 16 land all entering untapped does.
What about lands like Castles I hear you ask? These come in untapped for the most part and are activated at your discretion rather than as some kind of EtB effect. Certainly Castles are among the very best of the utility lands on offer and many remain in my cube. They are convenient and have minimal opportunity cost. The issue with these is tempo or power per mana if you prefer. You need 3 activations of Castle Vantress to see as many cards as Silundi Vision. You need to spend 40 mana on Castle Adenvale to make the same total stats as you get with Emeria's Call. In a world with endless time and mana the Castle cycle and lands with activated abilities in general outclass spell lands. In the real world of significant tempo pressures lands with activated abilities are one of the last places you sink mana into and as such rarely have that significant of an impact on games. I have activated less Castles in a year than I have cast spell lands in a few weeks. I have also laid more spell lands as lands than I have cast them which paints a pretty damning picture of the difference in relevance for these two groups of cards. Castles also occupy land slots in deck rather than the middle ground of spell lands meaning they generally wind up as a mild cost to your mana and thus consistency too.
Man lands are the closest in terms of power to spell lands but they are doing something different. Man lands main strength is resistance to removal as they evade all sorcery speed removal and mass removal when they want to while also typically dodging a lot of the rounded removal like Cast Out as it tends to state non-land. Manlands give you threat diversity and cause a headache for the control players. A lurking disincentive to planeswalkers is the other thing manlands do so well in cube. It is hard to compare them to the spell lands as a result given they perform these unique roles. Pound for pound the spell lands crush all the man lands in terms of stats on the board for mana cost. One combat step with a Treetop Village is 3 mana. You only have to pay that cost once for your Kazandu Mammoth and then you get to use it every combat there after until it is answered. It is also hitting rather harder for the most part too. So in terms of utility the man lands have the edge but in terms of tempo spell lands win out significantly. It doesn't matter much when both are clearly great groups of cards and feature heavily in cubes.
So there we have it, a load of technical reasons that spell lands are great. I didn't even look at the card design or flavour and they are big wins too. I love how they managed to pretty well sum up the colour pie for each colour in six spells. The only thing missing is a bounce effect in blue, and I guess some kind of Disenchant somewhere in the Selesnya area! They seem to have the power level fairly reasonable and consistent too. The quality of these cards and the clear thought gone into their design makes me really want to see a gold cycle summing up what the guilds are about. Being such a milestone in terms of card design and pushing the boundaries of the game they pretty much had to do a great job and my vote is that they achieved it. The only complaint I have is that they are a bit of a phaff to use in paper and feel a little like sac lands in terms of time messing around with cards added to a game! Luckily they are absolutely worth it. As will be all the extra wear these cards get in the years to come. The spell lands are a lot better than they look and do so in lots of subtle ways.