1: \begin{abstract}

2:   \protect

3:   Herbert Simon's classic rich-get-richer model is

4: one of the simplest empirically supported mechanisms

5: capable of generating heavy-tail size distributions for complex systems.

6: Simon argued analytically that a population of flavored

7: elements growing by either adding a novel element or randomly

8: replicating an existing one

9: would afford a distribution of group sizes with a power-law tail.

10: Here, we show that, in fact, Simon's model

11: does not produce a simple power law 

12: size distribution as the initial element

13: has a dominant first-mover advantage,

14: and will be overrepresented by a factor proportional

15: to the inverse of the innovation probability.

16: The first group's size discrepancy cannot be explained 

17: away as a transient of the model, 

18: and may therefore be many orders of

19: magnitude greater than expected.

20: We demonstrate how Simon's analysis was correct but incomplete, 

21: and expand our alternate analysis to

22: quantify the variability of long term rankings for all groups.

23: We find that the expected time for a first replication is infinite,

24: and show how an incipient

25: group must break the mechanism to improve their odds of success.

26: We present an example of citation counts for a specific field

27: that demonstrates a first-mover advantage consistent

28: with our revised view of the rich-get-richer mechanism.

29: Our findings call for a reexamination of preceding work 

30: invoking Simon's model and provide an expanded understanding going forward.

31: 

32:  

33: \end{abstract}

34: