There's some rough commonalities at the different point costs that can be used to get an idea of each how powerful existing abilities are. However, it's not a large pool of abilities, and the majority of them are 1 point, so it's going to be a very inexact process.
1 Point - Careful Spell, Distant Spell, Empowered Spell, Extended Spell, Subtle Spell, Transmuted Spell
This is the biggest category, and it has a lot going on. All of these abilities enhance a spell being cast in some way, but because of how varied these are, it's helpful to look at what they don't do. They don't interact with the action economy in any way, and they don't change any d20 rolls that affect a binary success/failure outcome of the spell.
Empower spell deserves special mention, since it increases damage dealt which borders on the action economy. But it doesn't increase the maximum damage beyond what's capable with just getting a lucky roll, and has a greater probable impact when rerolling an already bad roll, since you get more benefit from rerolling lower numbers.
- If your new metamagic changes how a spell is cast or it's features without affecting the action economy, the success rate of the spell, it's probably a 1pt ability. Abilities that increase damage dealt may also be 1pt abilities, but I would be careful of any that increase maximum damage or are more effective than Empower, especially because Empower spell is allowed to be used in combination with other metamagic abilities.
2 Points - Quickened Spell, Seeking Spell
These abilities both directly impact the action economy, but with notable restrictions. Quickening a spell to be a bonus action is helpful, but casters are still limited by the restriction against casting a leveled spell with both their action and bonus action, usually meaning one leveled spell per turn. This spell can still be used to increase overall spellcasting effect with spells that use your action every turn, like Eyebite or Sunbeam, but it won't let you double fireball. Similarly, Seeking Spell offers the chance to turned a missed attack into a hit, possibly saving an action from being wasted. But it's only helpful if you were going to miss, so it doesn't increase the maximum output on a turn, only saves you from bad turns and avoids wasting spell slots.
- If your new metamagic feature interacts with the action economy, allows casters to take more actions or avoid wasting actions and resources, it might be a 2 point metamagic. But interacting with the action economy still has significant restrictions even at this level.
3 Points - Heightened Spell
It's hard to judge what a 3-pt metamagic should be off a single example. Heighten is similar in many ways to Seeking - both take the d20 roll determining success / failure on a spell and give the player an extra chance to succeed. So why does this cost more? What spells use saving throws vs attack rolls could be it's own entire question, so the short answer here will be that in general it's easier to boost your attack rolls than it is to decrease an enemies saving throw, so this effect costs more sorcery points.
- If your new metamagic significantly hampers an enemy's defense against your spells, it might be a 3 point ability. Looking at Seeking vs Heighten, this might also be a good place to put upgraded abilities. If your new metamagic does something similar to an existing ability but with a more powerful effect/less restrictions, sticking it at three could make sense.
