If every puzzle has to have a forced mate after perfect play, why are all of you non GM's here trying to post puzzles ?
I have seen some puzzles in the trainer, where the losing side could have made a different move avoiding the mate for now, but not in the long term. Is it not a good idea to see the best move, after your opponent didn't ? If you only get good at looking a bunch of best moves, you might miss something you could otherwise take advantage of.
The point is that a tactic should give an advantage after any reply.
The computer opponent doesn't always have to reply in the best way (sometimes it gets you to prove the less obvious win through inferior replies) but your move has to have a gain against best defence otherwise how can you select the best move?
You can argue that any move is the start of a tactic by your logic, because the opponent could make a mistake you capitalise on next turn.
The point is that a tactic should give an advantage after any reply.
The computer opponent doesn't always have to reply in the best way (sometimes it gets you to prove the less obvious win through inferior replies) but your move has to have a gain against best defence otherwise how can you select the best move?
You can argue that any move is the start of a tactic by your logic, because the opponent could make a mistake you capitalise on next turn.
Are the only tactics you ever see in a game as a result of play after really good moves ? No... most of the time someone has made a mistake and it has lent to the creation of tactical possibilities.
Once again I ask the question, should we only be looking for tactics after someone made a really good move ?
If this how anyone trying to post here is going to be treated I recommend that there be a rating limit...2200+...that way the play was always good enough to justify someone using tactics....or ....
I know what, I'll start telling my opponents...awe, you just messed up, I should't play that tactic as it appears to me, ripe and juicy for the taking....no...I wouldn't want to offend the wanna be moderators over at the tactic trainer forum....
Don't you have anything better to do ?......really?
I have an idea, how about letting the mod disapprove and all other comment,argument debate be held else where. The first puzzle I submitted was only not accepted because it was originally one move long. As I said before, just because someone makes a mistake, doesn't mean the tactics that can develop are any less tactical. The better the moves that are made by ones opponent, the closer each best reply is to being a drawish position. Mistakes are required for tactical opportunities. GO troll in a troll forum.
I have an idea, how about letting the mod disapprove and all other comment,argument debate be held else where. The first puzzle I submitted was only not accepted because it was originally one move long. As I said before, just because someone makes a mistake, doesn't mean the tactics that can develop are any less tactical. The better the moves that are made by ones opponent, the closer each best reply is to being a drawish position. Mistakes are required for tactical opportunities. GO troll in a troll forum.
Look, if I make a random move in TT and the comp makes a stupid blunder like the puzzles you posted, that isn't TT at all. Mistakes ARE essential to tactics, but not in the process of a TT puzzle tactic.
Why does it matter when you take advantage of them or when they happen ? Generally when tactical opportunities arise, it is due to a mistake, what is wrong with showing the mistake and how to capitalize on it ? Why do you need to give the perception the position arose from good or mistake free play ? You are the one that isnt making any sense. The purpose of tactics trainer puzzles is to learn to recognize the mistakes and how to execute the tactics most efficiently afterward...
nameno1had,
I'm currently without a computer engine to help you, but looking at your smothered mate submission, you need to ensure the first solution move of 2.Nf5 is the best move available according to a computer. You might need white's dark bishop on h4 to pin the e7 knight, the e4 pawn on e3, and have 2.Nxf5 (ie.white gains material of some sort). The trouble is that your first move 2.Nxf5 should be more difficult than 3.Nd6# otherwise you end up with a low rated problem with a very low pass rate (something we try to avoid). The puzzle must start with a move 1... that precedes 2.Nf5 or 2.Nxf5. Black's next move 2...c4?? or other move needs to become something that saves black from losing material (eg.a rook) rather than gaining material (ie.white's bishop).
I always try to fix a problem submitted that doesn't fit the requirements, so unfortunately you are faced with trying to fix a problem I wasn't able to fix (that includes making the first move difficult), although I tend to try the hardest when I am trying to save a particularly beautiful tactic.
Have a look at the example of a smothered mate I gave you.
http://www.chess.com/tactics/server?id=317781
You will notice with an engine that black does not have to allow smothered mate, but that if he doesn't he will lose a lot of material.
This is sometimes allowable in TT because it forces white to prove that black had lost the original material. I hoped it might give you some ideas.
Achieving the above, to rescue your problem is difficult and time consuming. In the same time you could analyze 2-3 of your games with a computer, learning at the same time, and probably come up with several tactics.
nameno1had,
I'm currently without a computer engine to help you, but looking at your smothered mate submission, you need to ensure the first solution move of 2.Nf5 is the best move available according to a computer. You might need white's dark bishop on h4 to pin the e7 knight, the e4 pawn on e3, and have 2.Nxf5 (ie.white gains material of some sort). The trouble is that your first move 2.Nxf5 should be more difficult than 3.Nd6# otherwise you end up with a low rated problem with a very low pass rate (something we try to avoid). The puzzle must start with a move 1... that precedes 2.Nf5 or 2.Nxf5. Black's next move 2...c4?? or other move needs to become something that saves black from losing material (eg.a rook) rather than gaining material (ie.white's bishop).
I always try to fix a problem submitted that doesn't fit the requirements, so unfortunately you are faced with trying to fix a problem I wasn't able to fix (that includes making the first move difficult), although I tend to try the hardest when I am trying to save a particularly beautiful tactic.
Have a look at the example of a smothered mate I gave you.
http://www.chess.com/tactics/server?id=317781
You will notice with an engine that black does not have to allow smothered mate, but that if he doesn't he will lose a lot of material.
This is sometimes allowable in TT because it forces white to prove that black had lost the original material. I hoped it might give you some ideas.
I will double check with an engine and see what happens...
Think of a puzzle where you make a random move, and then your opponent makes horrible mistakes. Now that isn't a good tactics puzzle, is it?
What is random about strategically placing your pieces to increase pressure and gain space ? I am growing rather tired of you insulting my intelligence...
or simply escape being attacked ?
End of discussion. I am now communicating with nameno1had via messages instead of this thread. Everyone please stop filling up this thread with non-submissions.
If someone has a puzzle to submit, this would be a really great point to do so. 
Here is a puzzle for submission from one of my games. Black to move.
Very nice Deadey. Re2 is a nice move, and I really cannot see anything wrong with it.
The point is that a tactic should give an advantage after any reply.
The computer opponent doesn't always have to reply in the best way (sometimes it gets you to prove the less obvious win through inferior replies) but your move has to have a gain against best defence otherwise how can you select the best move?
You can argue that any move is the start of a tactic by your logic, because the opponent could make a mistake you capitalise on next turn.
Are the only tactics you ever see in a game as a result of play after really good moves ? No... most of the time someone has made a mistake and it has lent to the creation of tactical possibilities.
Once again I ask the question, should we only be looking for tactics after someone made a really good move ?
If this how anyone trying to post here is going to be treated I recommend that there be a rating limit...2200+...that way the play was always good enough to justify someone using tactics....or ....
I know what, I'll start telling my opponents...awe, you just messed up, I should't play that tactic as it appears to me, ripe and juicy for the taking....no...I wouldn't want to offend the wanna be moderators over at the tactic trainer forum....
Don't you have anything better to do ?......really?
It's clear you didn't understand my post.
Page32 submissions
Thankyou nyLsel for an amazing 185th puzzle http://www.chess.com/tactics/server?id=317597
Thankyou negativeb, for the 186th puzzle http://www.chess.com/tactics/server?id=374935. Unfortunately the next puzzle is ambiguous and the third too easy.
ThelronDuke, unfortunately Kf2 (mate in 3) is an alternative to Rb1 (mate in 2) making the puzzle ambiguous.
Samlong1989, thankyou for the 187th puzzle http://www.chess.com/tactics/server?id=475715 and 188th puzzle http://www.chess.com/tactics/server?id=415003 , unfortunately puzzle1 is ambiguous 1…Nxa2 (mate in 6).
Thankyou Player5-VERY_HARD, for the 189th puzzle http://www.chess.com/tactics/server?id=113041
wbame, thankyou for the 190th puzzle http://www.chess.com/tactics/server?id=417825 it had to start after the difficult moves because of alternative king moves making the puzzle ambiguous.
Thankyou Metaknight251, for the 191st puzzle http://www.chess.com/tactics/server?id=453153
cflux, unfortunately the puzzle is ambiguous and easy.
the_knife, for the 192nd puzzle http://www.chess.com/tactics/server?id=419267 , 193rd puzzle http://www.chess.com/tactics/server?id=245079 , 194th puzzle http://www.chess.com/tactics/server?id=398967 , 195th puzzle http://www.chess.com/tactics/server?id=378769
e4_guy, which moves do the puzzles start at ? (computer move first), and have you checked them with a computer to see if they fulfill the criteria in the first post of this thread ?
I've learnt how to use my chess computer better, so I've edited the first post again, to make it easier for people to submit a puzzle.
Interesting endgame tactic
Interesting endgame tactic
Of course it is far superior to tire the enemy king out before killing him, rather than killing him immediately with 31...Rc1# 
even if the a-pawn was a knight your idea would still not work because of 32.Kd1, but instead 31...Ra1 would win
You must use a powerful chess computer to ensure:
1)Each best move is at least +2.5 better than the next best move
...
examples allowed :
solution move +2.6, next best move +0.2
katonas, I was just looking at the initial post again and I think it contradicts itself. of course very marginal but are you saying that +2.6 vs +0.2 is allowed despite being only a 2.4 point difference between moves
pedantic I know but had to ask
Here, this fixes it once and for all (I hope, I need to figure out how to use my engine better I guess).
Yeah post 670 is a helpmate.
Does white get anything at all if 1...c6?