I think this is broken as all hell as it is, can I get some balance suggestions?

My suggestion would be to remove the ability to cast it from your graveyard.

Then it's a strictly worse Lightning Bolt - There has to be a way to make it work. Maybe change targeting to player only or make it 2 damage, not 3??

To be honest, it would be pretty powerful if it only dealt 1 damage. The only drawback would be that playing 4 would be a waste.

I suggest increasing the cost to . It's still good, dealing, on average, 3 damage for every six mana spent on it. Makes counterspells and discard seem silly, too. But at , there are going to be plenty of turns where it won't be to your advantage to activate this spell... better to put a 3/3 on the table, then potentially kill a 3/3. But as soon as the business spells are done, this card just keeps piling in more burn.

As an aside: Wizards current design models would probably cost this at . Partly because removal isn't as good as it used to be, and partly because Wizards hates to print coin flip cards that are economical. The last thing they want is for a tournament to be decided by a coin flip. It tends to undermine Magic as a 'game of skill.'

That might be why Link made his comment up there. There are many players who think coin flip cards should be strictly worse than the best options. I don't agree with that thought process... but I also don't choose to design many coin flip cards for this exact conundrum.

How about "Deal 2 damage to target creature or player. Flip a coin. If you win, return ~ to your hand after it resolves."?

That's still really good because it's multiple shocks, maybe it should be 1 dmg, but at least it's not multiple lightning bolts.

Or the flip could trigger in upkeep and be "return or exile". Or, indeed, you could flip when you first cast it to get the damage, but that would be quite feel-bad even if the card were good.

Mmm, damage should probably be unconditional. Flips should usually be pure upside even when you lose.

I wonder if this could be 2 to target if you win the flip; and 3 to target opponent if you don't?

As for how good it is... the once a turn limit is kinda vital; otherwise I'd just dump all my mana into this every turn - it's more than 50% better than fireball. Even at 2 damage it's slightly better.

I said what I said because a repeatable Lightning Bolt for us insanely powerful, even with the chance of failure.

I'm thinking: What if it exiled itself after a sucessful bolting? That combined with reducing the damage/making it unconditional as Vitenka said. could probably work.

Then what's the point?

Exiles itself from successful bolting if you cast it from a graveyard, of course. This is a pretty cool concept and I want to make it work.

So 50% of the time, it's a sorcery speed lightning bolt. 25% of the time, it costs . 12.5% of the time, it costs . 6.25% of the time, it costs . Etc, etc.

Reasonable. Still has its benefits in that it is strong against counters and discard. Bothers me that that's still strictly worse than Bolt... but bolt shouldn't be the standard. This compares well to Lightning Strike which has been the standard for a few years now. Also, a little unpredictable, but inevitable, so it escapes the 'game of chance' argument. Sounds fine.

This is not strictly worse than Lightning Bolt. Not at all.

I still prefer my version (cast automatically, do 1 or 2 damage, then 50/50 recur to hand or exile). My reasons:

• Most players will hate not knowing if they get a result the first time you cast it
• Most players will hate trying repeatedly and failing
• If it only goes off once as Jebediah's latest suggestion, it feels just less good than a spell you don't have to wait for.
• If you can try repeatedly, as in the current card text, it's way too powerful for a long game, where even every other turn, a free bolt will inevitably break stalls (especially if it can hit both players and creatures).

And also, yes, the average strength of a cheap burn spell should be about Shock or Lightning Strike, with Lightning Bolt as an upper end, and expensive red burn for limited at the lower end.

It's always a nice bonus if you can make a card equivalent to shock that's not strictly worse than Lightning Bolt, because it means it may sometimes be used in formats where people can use Bolt. But that's very much a secondary consideration, it's more important to get the strength right.