wickeddarkman

I've been through this deck in the past :D
For practical reasons you should play 2 hermetic study and 2 psionic gift to avoid cards like surgical extraction which will ruin the combo.

Try to have 5 of each relevant cards (5 crabs, 5 ping-enchantments, 5 draw enchantments and so on) which will increase the speed of you comboing with it.

I can tell you that my deck ended up being UG, because green adds deathtouch, so instead of placing a ping on the crab, you can place it on a deathtoucher, which will enable you to kill very large creatures with a single ping. Evasion is also important for curiosity, which is why I added river boa and marsh boa in the sideboard. Sigil of sleep is best on the crab, but you may want to place it on something unblockable as well if an opponent has pithing needle or phyrexian revoker.

How fast can you build up delirium?

Heh, I tend to detail my decks so much that few respond to them, and those who does I bark away with comments on how they don't understand the grand deck at all, so detailing CAN be overdone :D

Actually just try out a single archive trap :D
I've argued this with several others, but archive trap makes people react in a certain way, so if they see a single trap during game one, they will assume you have more, and try not to use fetches. So by playing a single trap you have a high chance of it occuring during the opening hand, but won't end up with too many dead cards in hand, and you slow down your opponent by fear of the trap, without really relying on it.

I've seen someone use overwhelming denial with surgical extraction, which migh be something you could imitate.

Also, you havent enough blockers/removal to support the planeswalkers. (you could cut 2 lands to gain 2 ensnaring bridge)

I'm currently trying to find a deck which could use authority of the counsil and mesmeric orb together to seriously stall/mill an opponent, and your deck seems to fit that theme due to peremiter captain and fog bank as well as the general theme...
http://www.mtgvault.com/wickeddarkman/decks/mesmericauthority-mill/

your midnight guard could also need to be enchanted with evanescent intellect, though that migh also need spellskite to protect midnight guard.

Also get 9 fetchlands for the hedron crab. Most milldecks that have gotten top 8 had 9 fetchlands in common.

Regarding your sideboard Traumatize:
You only want to be playing as much as two traumatize in your deck, because it does not pay out to play more than 1. In any match you will cast your first traumatize on turn 5. At that turn your opponent will have drawn 11-12 cards depending on who starts the game. You will mill about 24-25 cards from the opponent, leaving the same amount in their decks. They get 1 turn more ending up on 23-24 cards.
Playing a second traumatize will rob them of 11-12 cards, leaving the same amount. They then draw and have 10-11 cards which can be removed by either archive trap, mind funeral or glimpse the unthinkable, but not by a third traumatize.

Traumatize:
You only want to be playing as much as two traumatize in your deck, because it does not pay out to play more than 1. In any match you will cast your first traumatize on turn 5. At that turn your opponent will have drawn 11-12 cards depending on who starts the game. You will mill about 24-25 cards from the opponent, leaving the same amount in their decks. They get 1 turn more ending up on 23-24 cards.
Playing a second traumatize will rob them of 11-12 cards, leaving the same amount. They then draw and have 10-11 cards which can be removed by either archive trap, mind funeral or glimpse the unthinkable, but not by a third traumatize.

I must admit that I really like some of your innovative picks in this deck.

overwhelming denial backed by surgical extraction may have some really powerfull potential.
telepathy and inquisition also seems like a nice setup.

Here's some suggestions:

Lantern of insight might be better than telepathy if you write down what your opponents draw, and you may exploit this a lot more than telepathy with hedron crab and fetchlands. (Plus, lantern is colorless)

I usually test a lot against stuff like infect, affinity and deat's zoo, and against these I use 3 ensnaring bridge and 4 sage of epityr.
Sage of epityr may allow you to put just the right cards on your top to gain an early edge with what you need at the early game, though the only cards I can see removed to make room for it is 4 ancestral vision or possibly 2 dismember. Visions may be the better draw-engine, but sage of epityr can be used as canon fodder/blocker.

You may also opt to replace the two dispel with 2 collective brutality. Brutality helps an early ensnaring bridge, stalls an attack and may let you translate a deadhand archive traps into 2 life. (Archive trap is often a dead card against pro's and hinders your bridge)

Jace is also vulnerable to attack, so getting the bridge or some blockers into play is essential to you if you want it to become a planeswalker, so you should really consider adding sage of epityr.

I've written a lot about archive trap actually weakening mill, you see, once you play a pro, and they identify your deck as mill, they simply play around their fetchlands, and your stuck with 4 dead cards in hand. Since you only got 12 cards that are used in other decks than mill, they WILL identify your deck at round 1 or two, and that means that even during game 1 there's a chance that you wont get to use archive trap. When I played against pro-testdecks and played around archive trap myself, I discovered that it paid of to use 1 in mainboard, and I'd have 2-3 in sideboard against decks that are really harmed by ghostquarter (anyone playing 18-20 lands, or less).

Depending on where you play, snapcaster mage IS NOT the best choice. Snappy is the better card if you play small local tournaments, but once you play in 100+ tournaments, augur of bolas is a much better answer. Don't ask me why this is so, but if you compare all larger tournaments where mill wins, augur of bolas was always in the top 8 against a very large playingfield.

shelldock isle is an okay way to avoid burning out too early, but it ultimatelly stalls your gameplan. Use only 1-2 (I prefer using just 1) or it will give you some heavy mana troubles.

Personally I've never understood the fascination with visions of beyond, even when using it alongside snapcaster.
I've seen some of the later more "modern" decklists starting to cut it, or cutting it down to 2. I've always found that I prefer to replace all 4 with 4 surgical extractions, which each mill at least 3 keycards from the opponents deck.

Murderous cut kills 1 creature, and most top8 designs winning at larger tournaments use 2 damnation instead.

Ashiok does see more and more topplaces in modern mill, so if you want to play it, start defending it with more creatures or more removal. (augur of bolas and hedron crab will be a start).

Here's some crazy ideas:
alter of the brood uses permanents, so if more of your mill was permanents (shriekhorn beats memory sluice for example) you'd mill even more, and you could also use echoing truth and hurkyll's to really throw down some cheap cards in a flash.

Oh, the link was lost?!

I looked at some related pages, and at one time I saw what looked like safe haven was errataed to work as saurian,
but now that I try to find it again, none of the errata seems to work that way. Maybe I was tired?

Anyways, mosswort bridge and other hideaway lands may be activated by specific creatures, so they might be something.

By the way, here's some dubious tech from legacy!
There might be something there for you to use. Theres a "brand"-land that does nearly the same as the saurian, though I don't know if it's legal!

I know the feeling :D I sometimes foresee what the paperstrips will produce, and there's no greater feeling than knowing that an advanced method agrees with you.

I've come along way from playing jokullhaups in a deck with 20 lands because wotch recommends 1/3 land :D

I will start up my hunt for the perfect mana again, because all the paperstrip stuff has proven to be far better than my computersimulations ever were.

I will play against all the moderndecks with only lands and "vanilla" creatures (Creatures with no abilities) and produce the best manasolution on an average against modern.

Once I've taken it as far as I can with as high a winrate as possible against all matchups I will release it upon an unsuspecting magic community.

Anyone will be able to exchange the creatures in the manasolution with their own dreamcards and in their own choice of colors, so nothing will ever be the same again :D

My method will be for everyone...
http://www.mtgvault.com/wickeddarkman/decks/project-naked-deck/

I mentioned it because you've been through a couple of designs now :D
Since you might still add/remove colors I found it might be usefull.

One trick I often use is to fix the mana as the last thing being fixed at all.
Treat all your lands as whatever you want them to be, then focus on the deck itself.
Once the deck can't be improved anymore you fix the mana.

That trick will allow you to experiment wildly in many directions before finding the best build.

The deck does at least have a "look", but it's clear that there is a lot of conflicts between the cards, but I think it's fixable.
I like risendeeps suggestion with spellskite/proteus staff.

I also think a fokus on moldgraf is worthy of a try, perhaps with faithless looting which can double as a searchcard now that there's no serum visions.

Another approach would be to let them choose the "best" creature, only to play a lowcost card that suddenly turn the best into the worst. (or you may have already set up something like æther spellbomb long before)

Got it. The key is to fill the deck with creatures that only let's you get the advantage:

Myojin of life's web is a good example. (you get to play all your biggies if you get it into play)
Moldgraf monstrosity is another. (you block their biggie or kill your biggie to get 2 biggies)
kruphix, god of the horizons. (If they choose it they get a 4/7, if you get it you can double your mana every second turn.)
worldbreaker (when it dies, you can use it's graveyard recurring)

Using creatures like this will allow you to cut down the flickering...
If your deck is the only deck to really gain something from playing any of the large critters, then you wont mind that the opponent has a large critter as well.

Some other cards to consider:
boldwyr heavyweights (If you get two of these with dubious you prosper most)
doomgape (You have lots of elves)
Dread defiler (only helps you)
supreme exemplar (if you can get some elementals? maybe. might not work)

Ah, but if all the creatures are large enough to win the game it shouldn't be a problem :D

I know they aren't so large, but chancellor of the tangle could get you some early elves into play and a 6/7 ain't that bad. Besides it's 10 cards we are talking about. If a third of your deck is big critters you will get an average of three biggies with the dubious. You could make it into a different ponza deck, where elves will be able to hardcast your stuff.

How about cutting the bounce effect and relying on big critters alone?
It's much easier to hit two large than it is to hit 50% a blink + 50% a large. Simple is always better...