After deregulating their electricity markets in the 90s, many countries adopted free market concepts, such as auctions, for wholesale trading to meet electricity demand at the lowest cost while preventing rationing. Achieving those efficiencies depends directly on the equilibrium behavior of profit-maximizing power producers and the auction market design. Current works characterizing producers' strategies do not consider production cost heterogeneity, dependencies in production quantities, or abstract from replicating the market design, thus limiting the scope to homogenous producers and lacking concentration implications.

In this paper, we model the day-ahead market as a static Cournot game model of delayed production, with firms committing to a desired production quantity and subsequently producing a feasible quantity. We classify renewable firms (e.g., solar farms) and conventional firms (e.g., coal plants) by their production cost and ability to fulfill their committed quantities with certainty to achieve production cost heterogeneity. We link renewable firms through arbitrarily dependent joint distributions of production capacity, underpinning the prevalence of concentration on their production decisions. Our main theorem specifies that with less concentration, a tradeoff between (1) total production and (2) reliability occurs. To balance this, we implement a constant per-unit penalty on the production shortfall/surplus, reducing the incentive to overcommit while maintaining increased production. Finally, when faced with stochastically dominated distributions, the equilibrium results in renewable firms committing more, showing that delayed markets can provide hedging benefits to producers facing uncertain supply.