As you all know by now the Type 3 format has two distinctive features. One is that the maximum number of each card you can run is 3 instead of 4. The other is that we control the banned/restricted list. I'm going to take a look at the type 3 format and discuss exactly what that 3-card limit means. Then I'm going to look at the types of decks most effected by the format.
This obviously means I'll be talking about consistency. Really to maximize your consistency you want the largest percent of your deck to do the same thing possible. This will allow you do that thing more often. There are different ways of maximizing this percentage. The first of course is to just play with less cards. You will have a greater chance of drawing a specific card if there are less other cards in your deck. A 60 card deck will certainly draw the single Yawgmoth's Will you have in there more than an 80 card deck. 1/60 is a bigger number than 1/80. Another way to maximize consistency is to run more copies each card in the deck. A 60 card deck will certainly draw a card more often if it runs 4 copies of a card than if it runs only a single copy. Once again 4/60 is a bigger number than 1/60. Another way of maximizing consistency is to run tutors. Randy Buehler's Pro Tour Chicago deck from 2000 ran 4 copies of both Necropotence and Demonic Consultation. This basically makes the deck act like it has 8 copies of Necropotence. Once again 8/60 is a bigger number than 4/60. Similarly just drawing extra cards effectively reduces the size of your deck which increases your consistency. The last way to maximize deck consistency is by running multiple cards that basically do the same thing. This would be something like running 4 Lightning Bolts, 4 Incinerates, and 4 Chain Lightnings in your burn deck. While these are functionally different cards they all pretty much do the same thing (3 damage). Once again 12/60 is a bigger number than 4/60. I will admit that there are plenty of cards out there that are unique enough that there really isn't anything similar enough to use as redundant copies but most aggro decks and many control decks use this method of consistency maximizing.
So it is obvious to see that reducing the card limit to 3 of each card rather than 4 makes it harder to maximize consistency. I'm now going to look at each major deck type and the effect this loss of consistency has on them. I'll ignore our banned/restricted lists for these examples since it could vary depending on the decks that are abusive.
Aggro:
This is what I feel is the least effected archtype. It is generally the most redundant archtype and as such will have plenty of similar cards to fill holes with. I'll use an example:
This is a very consistent and pretty fast red aggro deck. It also has a decent late game with Grim Lavamancer and Cursed Scroll. Now lets see what it would take if we tried to convert it to type 3. We can replace the two non-basic land slots with 2 more Mountains. The rest leaves us with 10 free slots. I'd like to keep the deck running as close to the same as possible. We'll add 3 Incinerate to replace the missing burn. I'll also add 3 Frostling and 2 Suq'Ata Lancer to fill the missing creature slots. The remaining two slots will be a Blistering Firecat and a Mountain since the mana curve went up slightly. The deck now looks like this:
How has the performance of the deck changed? Not really much at all. It is about 1/3 of a turn slower but just as consistent as ever. Type 3 has have a very minimal effect on it. If aggro was strong in a normal environment it would continue to be strong in type 3.
Control:
At first you may think this archtype would be heavily effected by the lower consistency in type 3. However it actually isn't that bad. There are enough substitutes that the effect is still pretty minimal. It is more effected than aggro but not by much. I'll use a Draw Go deck as the example.
To make it type 3 legal we pull the extra cards and have 10 slots to fill. I'll add 2 Blinkmoth Nexus for the land slots. This keeps the man-land count at 8. To fill the 4 missing cards in the countermagic slot I'll add 3 Daze and a Foil. The last 4 slots can become a Powder Keg and 3 Serum Visions. The deck now looks like this:
How does it stack up to the original. Once again it isn't hurt quite as much as you'd expect. The most missed spells are the extra copies of Force of Will and Disk. Both have pseudo replacements though both are of a much lower power level. Everything else is easily replaceable. They may even games where this deck may perform better due to the extra free countermagic even if the deck has less hard counters. This deck is still quite the powerhouse and would make most of you very unhappy to see me play.
Combo:
Here is where we see the effect of type 3. Combo takes a huge hit to both consistency and speed. The reason is that except in very rare cases a combo deck wants to find a few specific cards as fast as possible. Not only does type 3 remove a copy of each of those cards it also removes a copy of each of the cards used to find them. Finding redundant replacements for creatures or removal is easy. Tutors tend to make a much bigger drop in power level when you have to run the second rate options. Sam has explained that Type 3 was originally created because his group was very annoyed that the environment was dominated by combo. While I wasn't around as much back then I'd be willing to be this was when type 2 was Tempest and Urza block, which was the most dominant combo ever was in type 2. I'd try to use another example here to try and convert a combo deck to type 3 but I really suck at combo and would likely start with a combo deck that is already terrible. If I used a proven tournament combo deck I'd probably ruin it just from lack of understanding. I do think it should be obvious enough that it would not stand up to the changes near as well as control or aggro.
Conclusion:
So we have concluded that the only archtype that is hit hard by the 3-card limit is combo. Type 3 also has a few innocent bystander cards that are now quite terrible because of the limit. Accumulated Knowledge goes from being an excellent card drawing spell to being quite terrible. Similar things can be said of the Oddessy burst cycle and Kindle. It also broke a few cards such as Jester's Cap, which we even had to eventually restrict. The fact that it can remove all copies of a card rather than leave one make it a much more powerful weapon (or at least more annoying). What I think really suffers though is any deck based around a single card. I'm not talking about combo decks but stupid things like Chance Encounter decks or something based around other dumb cards like Shared Fate or Timesifter. We started playing type 3 because we wanted to weaken the environment to allow stupid decks to be played. Since our environment had trouble not with combo but with control I don't think it was the correct call. After all I think the type 3 environment made control dominate even more. It certainly seem to win every week regardless of who played it. This was because the stupid decks were hurt by the 3-card limit more than the dominating control decks.
I don't want to sound completely negative about type 3. It had one feature that made me really like it. That was the fact that we controlled the banned/restricted list. This is what I think made the format fun. If we were to start playing a constructed format again it probably should be either an existing sanctioned format or a new home-brewed format with our own banned/restricted lists but with a 4 of each card limit.