So, the thing about Lorwyn is that they had a ton of great ideas about overlapping tribes (Benthicore, Bog-Strider Ash) and about including Changelings as glue, but the problem with how they made "overlapping" cards is that they don't serve in their typeline tribe. Take the Bog-Strider Ash I mentioned. It's not really a Treefolk, it's just a Goblin that doesn't benefit from other Goblins. So instead of being a supporter and a beneficiary for the same tribe like Boggart Mob may be, it's just a supporter for one tribe and a beneficiary of a different one, diluting it. Then there's the outdated hate cards, like the Bog-Strider, which explicitly weaken other decks pretty much by chance.
The whole race-class thing is, in theory, great, but there just aren't that many reasons to care about that class. Take soldiers, for instance. (This was my main deck back at the time, for kitchen-table Magic, so it's what I'm most familiar with.) On the surface, they look great: there are 28 Soldier creature cards, and only 12 of them are mono-White! The payoffs are all in White, so they're focused in that color but found in all colors that aren't Green. However, there's no reason to look for Soldier cards. These four cards are the entirety of the payoffs.
The thing is, they're all entirely in one of four sets, and one is either a Gray Ogre or a Serra Angel based on a coin flip, which is way too big a delta for me...maybe? If there was a top-of-the-library theme then that might be a cool payoff, but it's either way undercosted or never going to make your deck. The Tactician is good, but again suffers from being great or permanently stranded in the sideboard. Then, when you look at the Captain and the Banneret, two of the most consistent cards, there are only 3 Soldiers that cost 4 and 1 that costs 5! So you're not cheating much mana with either of the Captain and the Banneret. This problem shows up in most of the tribes. Compound this with the fact that most everyone wants Changelings and the fact that Champion directly hurts the ability to make tribes work, and apart from Faeries (which were actually quite good!), tribal effects just didn't come together. I think of Faeries as the example that proves the rule, in part because while tribes interacted at the level of the typeline, they didn't do a great job of being mechanically unified.
The effects were great, but they didn't match up well to what the tribes wanted and the support just wasn't there. Plus, the good glue was too generic. I think part of what you want for tribal sets is to have cards that will be fought over by, say, players in two to four separate archetypes (not everyone, but not just guaranteed to table to the tribal player the way Field Marshal would be).
edit: I cannot spell today.
The whole race-class thing is, in theory, great, but there just aren't that many reasons to care about that class. Take soldiers, for instance. (This was my main deck back at the time, for kitchen-table Magic, so it's what I'm most familiar with.) On the surface, they look great: there are 28 Soldier creature cards, and only 12 of them are mono-White! The payoffs are all in White, so they're focused in that color but found in all colors that aren't Green. However, there's no reason to look for Soldier cards. These four cards are the entirety of the payoffs.
The thing is, they're all entirely in one of four sets, and one is either a Gray Ogre or a Serra Angel based on a coin flip, which is way too big a delta for me...maybe? If there was a top-of-the-library theme then that might be a cool payoff, but it's either way undercosted or never going to make your deck. The Tactician is good, but again suffers from being great or permanently stranded in the sideboard. Then, when you look at the Captain and the Banneret, two of the most consistent cards, there are only 3 Soldiers that cost 4 and 1 that costs 5! So you're not cheating much mana with either of the Captain and the Banneret. This problem shows up in most of the tribes. Compound this with the fact that most everyone wants Changelings and the fact that Champion directly hurts the ability to make tribes work, and apart from Faeries (which were actually quite good!), tribal effects just didn't come together. I think of Faeries as the example that proves the rule, in part because while tribes interacted at the level of the typeline, they didn't do a great job of being mechanically unified.
The effects were great, but they didn't match up well to what the tribes wanted and the support just wasn't there. Plus, the good glue was too generic. I think part of what you want for tribal sets is to have cards that will be fought over by, say, players in two to four separate archetypes (not everyone, but not just guaranteed to table to the tribal player the way Field Marshal would be).
edit: I cannot spell today.
I hope they will do this more often but not so often that it becomes predictable.
