What are your thoughts on these chess engines?

CRITTER 1.4

HOUDINI 1.5A

STOCKFISH 4

STOCKFISH 2.2.2

STRELKA 5.5

FIREBIRD 1.1

GULL 2

BOUQUET 1.6

DEEPSAROS 2.3J

DEEPSAROS 3.0

KOMODO 3

TOGA 2

FRITZ 10

I am looking for advise on which chess engine/engines to use to analyze my games. I have done some research and downloaded a few chess engines. I wrote out a tournament spreadsheet and made the engines face each other. So I do have a pretty good understanding about how one of these chess engines correlates to the other in terms of overall strength. But what I don't know is which engine to use for what kind of analysis. I know the top three will likely be good for any tactical position but what about positional/strategic positions? What are your thoughts on these engines? Would any of these be good for analyzing a game for positional value?

For positional/strategic positions, the engines could suggest moves, but they won't be able to explain the reasoning behind the moves. That's one thing chess programs can't do yet.

Short answer : none of them.

Detailed answer : as engines, no matter how strong, are unable to explain the reasons behind the moves they make, the only use they have is to blunder-check your lines. All they can say is "this is better because I deem it better via brutal calculation".

With a simplified example, imagine that you have downloaded Houshreddybka 100.0 that can find the best move in any position immediately. After inputing 1.e4 d5, he tells you the best move is 2.exd5 that wins whereas 2.e5 just draws and 2.d3 loses - but you can make no use of that information : you can go down the lines and find that after 500 moves the evaluation is indeed this one, but whoever cares. However, if you give him 2.Qg4 he will tell you this is losing with the line 2...Bxg4, and this is useful to you if you overlooked that move (2...Bxg4), but because you can actually read the whole variation and understand that the final position is losing.

Analyse by yourself then try your lines with an engine of the same rating as you and then with an engine 200 elo above you.

If you aim at making progress in chess, analyze by yourself first (use your brain), then blunder-check with any chess engine.

If you want to understand positional chess, better go with annotated master games or analyze with a strong player, as chess engines can't explain strategy.

Good advice here.  Also be aware that in trying to compare chess engines, there are many variables that will change results.  For example, some engines may be able to use multiple cores of your CPU, others are single core.  Some engines are better at quick move time analysis, others may excel if you give them more time per move to analyze.  Also, note that engines use databases, especially for endgame analysis, and the quality of the database is independent of the engine itself.

As @hicetnunc said, do the work yourself, make notes of your choices or candidate moves, and then check with an engine.  Simply running an engine on a completed game does not help you learn to think as you need to do in OTB play.  Engines run lines more deeply than most humans and therefore their evaluation of a position may well be beyond practical limits for most of us.

Personnaly, I prefer Fritz 2013

Fritz does attempt to inject commentary with its analysis if you activate that.  I don't know how strong the Fritz 13 engine is compared to (for example) Houdini, but it's plenty strong enough.  I don't find the comments to be all that helpful.  It is also difficult to assess the strength of Fritz since the program is updated so frequently (currently on its 32nd update).  They don't tell you what is being updated - engine strength, bug fixes, copy protection, etc. - but it works for me.  Good program.

Look: it is an open secret which are the best chess engines ( http://www.computerchess.org.uk/ccrl/4040/ ). Houdini is undoubtedly the best, it can make full use of your hardware, it is capable of fast and deep search and my personal experience is that it has a masterful pawn game and very deep understanding of pawn structure.

Oh btw, I should step in and say that Houdini 3's endgame is a pain to watch, however. It is probably the best in all other areas, but I have seen him butcher endgames like a 1600 here wouldn't, in the last Candidates.

So if ever you have the bad idea to rely on computers for endgames, above all do not rely on this one.

This is all very important - if you are playing against another computer.

Or you are a certain bulgarian player and playing against a grandmaster. Also then Houdini prevails.

Alright let me see if I have this right and correct me if I'm wrong. None of the engines mentioned above play postional/strategic chess and in the event that any of these engines do suggest or plate a positional move is simply sheer luck and that it played or suggested the move solely because of the highest value and it had no idea the move necessary had any postional merits?

I don't think you can say that, exactly.  The algorithms are programmed by humans, and different engines have different strengths and emphasis.  As a general rule I think most engines rate positions by material balance, but what is a good line as seen 10 or more moves ahead is likely to have a lot of strategical or positional merit.  Remember that these engines are rated above 3000 and Magnus is "only" 2800+.

The problem is that the engines cannot tell you why a move is good except to show you the resulting lines and results of those lines.  They are not teachers, they are analyzers.

i recommend stockfish because it is very strong (#2 of all engines behind Houdini) and it is free.  I use stockfish with my Arena chess interface (also free).  

I have played kentaur chess at FICGS with Houdini and it is capable of making positional decisions. Once it was able to hold a position an exchange and a pawn down just because it could create a very solid pawn chain with an untouchable centralized knight in the middle of it. You can call the exchange and pawn sacrifice it made positional decisions and very wise and long term ones for that matter.

I would guess that these engines do not "think" but rather test lines by repetitive "brute force" until they find the best line.  Material exchanges and sacrifices are included in "test" lines and if the resulting position is favorable, it is played.  The final result may be a tactical brilliancy, but the computer did not think or plan, it simply tried lines until it found the strongest continuation.  If you limit the thinking time per move on these engines, or limit the number of plies, you'll get different results.  The more time to test variations they are given, the better the results, of course.

Positional understanding is not a mystical thing. It is positional understanding that having the bishop pair in an open position is a good thing. Both 'bishop pair' and 'open position' can be objectively measured and are quantifiable elements. So it is possible through the evaluation function to give positional understanding to engines. If you see twenty moves ahead with some positonal understanding you are capable of play which humans would call very strong positional play.

This is the best chess engine I have found...

I'm sure all of the engines mentioned by the OP have positional aspects to their evaluation functions. The user's problem is that this evaluation process is typically a black box. The various evaluation criteria (material count and various positional adjustments) go into the black box, and a final evaluation number gets spit out and presented to the user...This line is +1.24.  OK, fine, but that number doesn't tell the user whether the line is tactical, positional, or some combination of the two. It takes a strong, knowledgeable person to understand what the line is doing.