Apollo is playing against Nostradamus. During Apollo's upkeep, the Courser of Kruphix of Apollo is destroyed. However, after a long enough pause to allow further responses, he moves to his draw step, draws a card (which is still revealed), and then reaches for his library again. You, as a judge, are observing this match.

Now, it is almost guaranteed that Apollo is reaching for his library to reveal a card for the Courser of Kruphix that is no longer there. It is easy for you to make a remark along the lines of “the top card shouldn't be revealed anymore” to prevent an error from occurring.

However, according to the IPG, "Judges don’t stop play errors from occurring, but instead deal with errors that
have occurred "

I can understand why judges aren't expected to prevent errors: they are not able to predict the future. However, the IPG seems to actively forbid judges from preventing errors. What is the philosophy for that?

I see two reasons behind this, besides the mentioned inability to see the future.

a) If we prevent one error we have to prevent all errors and we just cannot since we cannot be everywhere at once (and we are also fallible)

b) If we cannot prevent all errors, we might appear to be favoring one (a few) player(s) over the other(s).

You should step in when you see an error occur. In this situation, you
have a Courser that is no longer on the battlefield, yet the top card is
still revealed. You should step in at that time and correct the error.
Once that is corrected (top car is facedown), if the player draws and
starts to reveal the top card again, you do not intervene until the mistake
occurs.

Shawn, are you saying that you would give a L@EC warning because Apollo did not explicitly turn the top card face down, even though the draw is the first thing that happens after the Courser dies? Or are you arguing that a judge can intervene in this specific scenario because technically an error occurred?

Regardless, I'm trying to use this as an example in order to discuss the philosophy behind a judge not intervening when he sees an error is about to occur. I am not necessarily interested in the details of this specific scenario.

Until the moment the error happens there is a chance the player can catch the error still.

Also if we were allowed to stop a game to stop an error do we still issue a penalty for the error?

You intervene to tell the player to turn the card down. No penalty is
needed.

We do not prevent mistakes. It is beyond our responsibilities. Mostly
because we are guessing what *might* happen.

Not resolving your Slaughter Pact trigger is also an error. I would hope no judge steps in when my opponent is about to make that error.

Shawn is right on with this. And really, that first paragraph in the IPG that Michel quoted part of? Taken as a whole, it lays out the philosophy:

Judges are neutral arbiters and enforcers of policy and rules. A judge shouldn’t intervene in a game unless he or she believes a rules violation has occurred, a player with a concern or question requests assistance, or the judge wishes to prevent a situation from escalating. Judges don’t stop play errors from occurring, but instead deal with errors that have occurred, penalize those who violate rules or policy, and promote fair play and sporting conduct by example and diplomacy. Judges may intervene to prevent or preempt errors occurring outside of a game

While watching some of yesterday's Top 8, I noticed Kevin Desprez illustrating perfectly what judges should do, and in a scenario very similar to Michel's - i.e., a Courser that's no longer there, and a card that gets revealed.

Kevin waited until it was actually an error, then stepped in, applied Looking At Extra Cards and shuffled the card into the random portion of the library. (Granted, some of that is my assumption based on what he did, and how he interacted; I couldn't actually hear him.)

Michel, you asked about the philosophy behind that. In order to remain neutral, and to avoid appearances of bias, and to allow players to make strategic as well as rules mistakes, we MUST avoid intervening until it's appropriate.

d:^D

In that case, was the actual error doing something else with the top card still revealed (since it was being seen without anyone having the right to see it), or was that let go and he stepped in when the next card was to be revealed (since people already really knew what it was, so why interfere earlier)?

“Manipulated”, in this context, means cards that have been legally
manipulated by the player – through scrying, Brainstorming, and so on. If
I forget that I no longer have a Courser in play, and flip over the top
card of my library, I may be “manipulating” that card in the dictionary
sense of the word, but not in the sense that the IPG means. So that
revealed card would be shuffled into the random portion of the library.

Can I get a clarification of when exactly L@EC applies? Is it the moment the Courser dies and the top card is not flipped back over? Even though both players know exactly which card it is? Or is it when the top card changes (eg draw step) and the next card is mistakenly revealed?

“A player looks at a card they were not entitled to see.” - continuing to see a card that was already revealed doesn't fit that definition.

Leaving your top card revealed after your Courser of Kruphix dies is, technically, a Game Rule Violation - but it's not worth assessing an infraction, just remind the player to turn it back face-down.

Once a new card is revealed - because you forgot your Courser is gone - that's L@EC.