Well, to be clear, i didn't intend to not add the card to the game only because it is overpowered for the cost. The numbers can be changed, but when making concepts, it would be great to have quite balanced powers. And this issue leads to topics of card changes when some cards would be called underpowered when they aren't.
It's not the problem when a pair of cards are more powerfull than others (Although i would still like to see a nerf of anthoria, her attack value should be drastically be reduced, so the legend keyword can be kept and she is still incompareable with lord of time which would be a nice change). The problem is when the overall power of cards increase, the game itself changes in a very drastical way, which leads to synergy decks being underpowered and only individual powerfull cards matter (See MTG's standard-format for that reasons, its almost a rare-only format, why are they even printing commons then?).
Skald is a great concept, but we could take away some work from the developers and add cards to this section with a proper cost.
And ell, you're right that there is no card that is exactly like Skald. But what isn't true is that it's incompareable. It needs some abstraktion though.
-The condition is quite simple met. In an soldier, rush or barbarian deck (which all are aggressive strategies) you are most time playing recruits-only cards. So in these decks, the condition for this card is met like in 80% of the cases. The rest of the cards only summon new cards ti kick in the opponents face or are adding recruits to the player.
-So this card needs nothing counter-intuitive or doesn't require much amount of planning, since it plays perfectly into what you want to do anyway (Except the legend part).
-This creates a situation in which the effect should not scale to high, since the risk/reward ratio is quite low.
-So we an pick a no condition card which givesa compareable amount of recources (I would say in avarage skald gives 10 recruits if played properly) and can adjust the cost on a card with a discount.
Enter magical steel:
http://arcomage.net/?location=Cards_details&card=50And now we directly see how the legend condition should be balanced: It shouldn't give more than 7 recruits (A 50% discount because legend cards are
rarer than mage cards, but both are equal counterintuitive)
The ceiling of skald is much higher but for this it still got a good variance and recruit cost, because of which it is compareable to a plain gain of the same average amount.
What just really hits it out of the borders are the 15 recruits for the legend card. Or the variance needs a hit with a cap.