I watched an intriguing talk on TED by Peter Donnelly, a statistician with a specialty in genes, that spawned some thinking about deck building. The talk it about 20min and I highly encourage you to watch it - but if you arn’t in the mood, let me spoil a little of it for you.
Peter talks about a simple coin-flipping exercise. One side of the room is interested in getting a combination of HTT (heads, tailss, tails), while the other half of the room is interested in getting the combination HTH. Statistically speaking, everyone believed that after a large enough data set, both combinations would have an equal chance of appearing. Additionally, Peter asked the question of which combination will appear first? Nearly everyone thought (as I did) that they would occur, on the average, at the same time in every test. This was wrong: HTT appeared first because of it’s pattern vs HTH. However, if you where looking for HTT and you have your first H and first T and then are delt a H you are effecitvly starting over; if you where looking for HTH and have HT a second T would be devastating, but a second H would not only complete your pattern but give you opporunity to start a new one.
Peter talked more about this and how it relates to his specialty, but for our purposes the point is made: when you stand back and look at the entire deck, one set of card choices may very well be equally powerful as another, but that doesn’t mean that it will produce results for you as fast or as concentrated as you might expect. This is profound. This explains why Kitchen Finks are better than, say, Nyxathid in a Death Cloud deck.

Kitchen Finks
Before Death Cloud: 3/2 creature and +2 Life vs X/X creature OR no creature
After Death Cloud: 2/1 creatue and +2 Life vs no creature OR Y/Y creature
Do you see how Kitchen Finks provides card advantage over Nyxathid? It pratically plays itself a 2nd time after the cloud goes off. Additionally, it is covering up two of cloud’s weaknesses (you loosing creatures and you loosing life). A 7/7 monster after a board wipe is scary, but at the cost of a Thoughtsieze/Distress followed by a Deathcloud THEN to draw Nyxathid… you are making things complicated for the sake of a big stompy that can be removed or mitigated easily (Path to Exile, Bitterblossom). Do you see the HTH vs HHT pattern?
The obvious pattern of choice with a card like Death Cloud is HTH. So, what else goes good with Deathcloud other than the obvious Tarmagoyf? Lets look at Deathgreeter first. Pretty mediocre when you first play it. You wouldn’t want to use it to block: basically this guy stops you from loosing most if not all of the life to cloud. But now we’re thinking.
Toshiro Umezawa is an interesting pick. At 3/3 while blocking, he can beat back smaller opponents. And if your deck is set up correctly, you will be slinging spells at the opponent (including instants that where in your hand just a second ago - read the order of the card). Getting warmer, but this guy requires a specially tooled Death Cloud deck compared to what we’re use to.

Dusk Urchins
Dusk Urchins has to be my favorite. You want to use him to block, and you can use him for a swing or two. This guy is good for one to two cards after Death Cloud. How much better does it get? That is a HTH if I’ve ever seen one.
Lastly, cards like Deadly Grub, Dregscape Zombie, Golgari Thug and Nether Traitor need to be mentioned. While not the biggest bang for your buck, all provide decent answers to the lack of creatures after the cloud goes off while providing potential beforehand.
Take this knowledge and grow, my young friend! When you start winning cash prizes, e-mail me and I’ll give you my address so I can collect my fair share.

