I have no data to support my comments, only knowledge.
As gaslight more or less stated, you need to know the order in which the bar extremes occur. Looking only at conventional data points, OHLC, is incomplete in this regard. Time would be a further criteria once the order of bar extremes is known.
Formation of most bars BEGINS as inside bars and morph from there. If/when an inside bar morphs into another type of bar, that bar can not return (never, as in 0% chance) to being an inside bar. A similar law applies to outside bars... once a bar has become an outside bar, it can not become (never, as in 0% chance) anything else: It will be an outside bar. To be thorough, discussion of Doji or other "identification" is not applicable here as those identifications are not bar formations. An outside bar or an inside bar or just about any other bar formation can be a doji, for instance.
As gaslight more or less stated, you need to know the order in which the bar extremes occur. Looking only at conventional data points, OHLC, is incomplete in this regard. Time would be a further criteria once the order of bar extremes is known.
Formation of most bars BEGINS as inside bars and morph from there. If/when an inside bar morphs into another type of bar, that bar can not return (never, as in 0% chance) to being an inside bar. A similar law applies to outside bars... once a bar has become an outside bar, it can not become (never, as in 0% chance) anything else: It will be an outside bar. To be thorough, discussion of Doji or other "identification" is not applicable here as those identifications are not bar formations. An outside bar or an inside bar or just about any other bar formation can be a doji, for instance.

