Should stalemating count as 3/4ths of a win?

They probably will stay the way they are.
In both FIDE and USCF. Different.
A player playing someone who doesn't know the rules ...
The player has two knights against his lone King.
Lone King resigns. The player takes the win with signed scoresheets.
Later - he finds out the rule. Appeals.
Too late.

Do not know what all here are thinking.

Myself loves MariasWhiteKnight.

That is not the best case because even under FIDE the player can get a draw by the 50 move rule and if there is any type of delay or increment the time will be available.

Unless you are talking about the player considering the already fallen flag to signal a loss.

Your smurf (me) says Draw.

@jetoba I didn't word my post quite right.
I'll fix it now.
The player with the lone king resigns.
Is what I meant.
I think you gathered that.

Regarding the 50 move rule - in the example given I'm talking about a common situation with a player short on time and then flagging out.
He has seconds left and has just lost his last material.
He doesn't even consider trying to blitz out 50 moves.
Or tries and fails. Then resigns/gives up when the other player calls 'flag'.

I should mention that there is a particular situation where a player can lose if the other player flags - and that is 'board position'.
For example:
Player calls 'mate' the other player calls 'flag' and the first player already is setting the checkmating piece down. Or its already stalemate for a draw.
The tournament director is supposed to go by the position on the board - whether checkmate or stalemate. Not the clock.
Many might say that calling 'mate' isn't good enough because it can't be proven or one player could claim he didn't hear - that he was 'concentrating' ...
or that it just isn't the rule.
I've seen these situations in live tournaments by the way.

What defines a "forced mate". Where do you draw the line? A mate in 3 or a mate in under 20? Either you include all mates, forced or not, or let an arbiter have the discretion to award a draw to any person reasonably up on material that flags. I see nothing special about the king + 2 knights vs king that makes it any more drawish than KNKB. As I said, in some positions just making the wrong move can get you mated:

Yet actual Endgame studies exist where the last move results in a KNKB but would be drawn under this sites version of USCF rules:

This site would immediately declare a draw before white could play ng6#, which is absurd. 99% chance white will see that checkmate at that move, gets stuck with a draw, yet .000000000001% chance 2 queens would lose to 1 pawn, yet the pawn wins if flagging. It's totally backward. At least FIDE keeps it consistent, 2 queens lose to 1 pawn the same way a king loses to 2 knights or 9 queens + 2 rooks lose to 1 knight. As long as a mate exists with those pieces, it's a win/loss. The only piece combinations that are a draw are KBKB same colors, KNKQ, KBKQ, and KBKR, for the side with more material, if they flag.

Because a knight can checkmate a rook, in fact black in this position should lose on time if he flags because this is possible:

"This site would immediately declare a draw before white could play ng6#, which is absurd"
Yes it is absurd.
But it kind of has to do it apparently because otherwise the rules would be inconsistent in the case of using computers.
Because a player might be playing for the draw that way.
They're apparently all different.
You could argue that they're all absurd.
FIDE - USCF - chess.com. ...
but they've got their reasons.
Things aren't always what they seem to be.

That's why any mating material left rule would be the simplest. KBKB same colors, KBKR, KBKQ, and KNKQ are the only piece combinations where the former side cannot mate the latter. Any other piece combinations that isn't insufficient mating material for one or both sides, the side that flags loses, simple to program.

You could consider that K+R versus K+R scored as a win against the flagged player who loses - is absurd even though it has helpmates.
Its essentially a drawn position.
How many types of position where flagging loses that are actually draws - are there?
Queen versus queen is usually a draw.
Many pawnless endings are a draw.

Queen plus knight versus Queen plus knight should be a draw but it could be easy for one side to blunder.
In some over the board tournaments USCF had a rule they may still have that if one player wants a draw to be adjudicated in a simplified position then the tournament director is to form an opinion as to whether a master could swindle a C player in the position and if he thinks the master might be able to do so then no draw.
Maybe FIDE has a similiar rule.

But you can't have something like that in an online chess site.
The computers just have to decide instantly.
And in some cases the players have to agree.
And absurdly - the computer won't give a draw in K+R verus K+R.

Isn't it reasonable to suppose that USCF and FIDE must both have a rule that either player can claim a draw there?
I've encountered some situations on chess.com where I pressed a draw button and the draw was instantaneous. Years ago.
Had the impression the computer did it. But wasn't sure.

Rook vs Rook endgame is winnable even without blundering the rook!

So why should white get a draw if he flags?

That's called insufficient losing chances. Not sure if FIDE has a similar rule but unlikely as it already has the "any possibility of checkmate rule", but you can argue that if a position is really dead drawn like KRKR, then both players should have no trouble holding the position even with 1 second on the clock. In other words they shouldn't flag in the first place. Not sure where they got "Master vs Class C" judgement decision from, sounds arbitrary. Maybe something like "against someone 1000 points higher" would make more sense IDK.

USCF No longer has insufficient losing chances either.

They have abolished it within the last 5 years. Now, if they are using an analog clock, tough luck. The assumption is that nobody will use an analog any more. You cannot use ILC with any clock that has a delay feature or an increment feature.

That depends. There also must be a legal sequence of moves that leads to mate for the side with time. In the following cases, it is a draw if Black is to move and his flag falls, despite the material, in FIDE:

Fide has lost the plot.

How so? FIDE has it right. USCF is the one that's bleeped up!

Although K+B+4P vs K+B+4P is sometimes drawn in USCF and FIDE even if one player flags. Special positions like White Pa4, Pc4, Pe4, Pg4, Bf3, Ke2 and Black Pa5, Pc5, Pe5, Pg5, Bf6, Ke7. Though Chess.com sees that K+B+4P can beat a lone K and awards a win.

In the three positions you gave USCF would also call them drawn, though not every TD/Arbiter knows it.

A forced mate is one where you can force a mate even if the opponent plays perfect moves.

For instance Black Kh8, Ph7, Pe6, Bg8 and White Ke1 Bd2. If White plays Bc3 then Blacks only legal response is e5 and Bxe5 is a forced mate. In FIDE and USCF if Black flags instead of playing e5 then White still wins. In Chess.com if Black flags instead of playing e5 then White K+B is considered insufficient and the game is drawn.

If the Black bishop is on f7 instead of g8 then in FIDE if Black flags it is still a loss (the game could go ... e5, Ke2 Bg8, Bxe5# while in USCF there is no longer a forced mate (there are helpmates but not forced mates) and USCF will call it a draw.