Abstracted Scoring

Games often determine a winner by some kind of scoring technique - usually something universal like victory points. Sometimes the VPs are hidden behind a screen, on the board itself, counted in cash, or hidden in plain sight.

How the scoring record is handled, matters a great deal. If it is something concrete that other players can constantly refer to, players can and will base their playing based on who is in the lead.

Scoring can be incomplete, due to unrevealed tricks or just not far enough in the game. As an example of not being far enough - in Agricola, each turn fixes some part of their farm up more. You can kind of keep up on how much they're missing and try and cut them off from the boars you think they need, but you can't accurately gauge their scoring potential. Likewise, some games reward sets of cards in play, and scoring things just on what you see before you doesn't address the unknown cards in your opponent's hidden information.

Abstracting the score tracking methods adds to that complexity.

Option 1 is the best place to be. You understand the scoring complexity, and you're already ahead.
Option 2 is second best position. You're not in first, but you understand how it can change. You can point out how someone else is winning.
Option 3 is great - you're in first, but you're not to the level of understanding how scoring works - someone could pull ahead, and you can't explain to other players when other players overtake you.
Option 4 is the worst - you're not necessarily sure where you are in scoring, but you're not in first.

Abstract scoring gives the Inexperienced Players a double hit - they're less sure how to win, and can't tell how well others are doing. (You can also apply a middle range, called the "Distracted experienced player" - knowing the rules, but distracted by the company of friends, girlfriend/boyfriend, teaching the game, etc.)

This isn't to say, abstract scoring is always bad - if a game allows for a quick advancement into being an experienced player, it evens things out a bit. But a dense, meaty game where it'll take 4-5 hours to play, adding abstract/hidden scoring creates a huge inequity gap between experienced and inexperienced players.

Example:
Through the Ages has a scoring track. It is accurate up until Age 3, where Wars and Goals kick in...new players could play the Advanced Game through Age 2 often, but Age 3 would always kill them against experienced players. Mitigation: a list of possible goals, and a thorough example of wars. (I usually recommend this at the end of Age 1.)

Abstracted scoring can also be a good mechanic for your game if it includes hidden elements. Experienced players might be able to guess what kind of information is hidden, but not the specific qualities of the hidden data. Again, experience will provide an edge, but in this case, they're stuck guessing how well the inexperienced player has caught on. (You can't guess their actions, because they haven't found the 'ideal' experienced playing style.)

Enjoy some games this weekend!