diff --git a/paper/figures/WA_poc_seats_chartsystem_compare_pared.png b/paper/figures/WA_poc_seats_chartsystem_compare_pared.png
new file mode 100644
index 0000000..91d33d0
Binary files /dev/null and b/paper/figures/WA_poc_seats_chartsystem_compare_pared.png differ
diff --git a/paper/paper.md b/paper/paper.md
index ff80a09..b226b71 100644
--- a/paper/paper.md
+++ b/paper/paper.md
@@ -131,7 +131,11 @@ See generally
 references.
 
 Reform advocates also need to describe voting mechanisms and their
-likely outcomes effectively to members of their communities. `VoteKit`
+likely outcomes effectively to members of their communities. The end-to-end pipeline provided by `VoteKit` allows advocates to toggle different system settings and compare expected outcomes.
+
+![A comparison of a variety of electoral systems and their affect on minority representation in a case study of the Washington state legislature [@washington_leg]. \label{fig:WA_comparison}](./figures/WA_poc_seats_chartsystem_compare_pared.png){width=100%}
+
+<!-- `VoteKit`
 includes a variety of metrics and visualizations intended to facilitate
 this.
 
@@ -144,7 +148,7 @@ shows head-to-head margins between candidates. For instance, 282 more
 voters ranked $W_2>C_1$ than $C_1>W_2$ in the preference profile used to
 make the figure. Note that $W_1$, who has the most first-place votes by
 far, is also preferred head-to-head over all alternatives, making them
-the *Condorcet candidate* in this election.
+the *Condorcet candidate* in this election. -->
 
 ## Area of need: Resources for research
 
diff --git a/paper/paper.tex b/paper/paper.tex
index d116cf3..71e55fa 100644
--- a/paper/paper.tex
+++ b/paper/paper.tex
@@ -90,36 +90,40 @@ \subsection{Area of need: Comparison and communication}
 \end{itemize}
 See generally \cite{electoralhandbook,STV,Borda,TopTwo,SequentialRCV} for references.
 
-Reform advocates also need to describe voting mechanisms and their likely outcomes effectively to members of their communities. 
-\VK includes a variety of metrics and visualizations intended to facilitate this. 
-
-\begin{figure}[bht!] \hspace{-.5in}
-\begin{tikzpicture}
+Reform advocates also need to describe voting mechanisms and their
+likely outcomes effectively to members of their communities. The end-to-end pipeline provided by \VK allows advocates to toggle different system settings and compare expected outcomes.
+\begin{figure}[bht!]
+    \includegraphics[options]{./figures/WA_poc_seats_chartsystem_compare_pared.png}
+    \caption{A comparison of a variety of electoral systems and their affect on minority representation in a case study of the Washington state legislature \cite{washington_leg}.}
+    \label{fig:comparison}
+\end{figure}
+% \begin{figure}[bht!] \hspace{-.5in}
+% \begin{tikzpicture}
 
-\node at (2.5,-5.2) {\includegraphics[height=2.5in]{../figures/pwc_graph.png}};
-\node at (2.5,-8.2) {\bf Pairwise Comparison Graph};
+% \node at (2.5,-5.2) {\includegraphics[height=2.5in]{../figures/pwc_graph.png}};
+% \node at (2.5,-8.2) {\bf Pairwise Comparison Graph};
 
-\node at (10,-5.2) {\includegraphics[height=2in]{../figures/profile.png}};
-\node at (10,-8.2) {\bf Preference Profile};
+% \node at (10,-5.2) {\includegraphics[height=2in]{../figures/profile.png}};
+% \node at (10,-8.2) {\bf Preference Profile};
 
 
-\node at (0,0) {\includegraphics[width=2.5in]{../figures/fpv.png}};
-\node at (0,2.2) {\bf First-Place Votes};
-\node at (6,0) {\includegraphics[width=2.5in]{../figures/borda.png}};
-\node at (6,2.2) {\bf Borda Points};
-\node at (12,0) {\includegraphics[width=2.5in]{../figures/mentions.png}};
-\node at (12,2.2) {\bf Mentions};
+% \node at (0,0) {\includegraphics[width=2.5in]{../figures/fpv.png}};
+% \node at (0,2.2) {\bf First-Place Votes};
+% \node at (6,0) {\includegraphics[width=2.5in]{../figures/borda.png}};
+% \node at (6,2.2) {\bf Borda Points};
+% \node at (12,0) {\includegraphics[width=2.5in]{../figures/mentions.png}};
+% \node at (12,2.2) {\bf Mentions};
 
-\end{tikzpicture}
+% \end{tikzpicture}
 
- \caption{Four visualizations for the same synthetic four-candidate preference profile with 1000 voters.  The profile was generated with the CS model, which produces incomplete ballots at a realistic rate.}
- \label{fig: votekit summary stats}
-\end{figure}
+%  \caption{Four visualizations for the same synthetic four-candidate preference profile with 1000 voters.  The profile was generated with the CS model, which produces incomplete ballots at a realistic rate.}
+%  \label{fig: votekit summary stats}
+% \end{figure}
 
 
 
 
-For a given preference profile, basic \VK functions provide statistics and visuals for first-place votes, Borda count, and mention frequency, as well as head-to-head comparisons (see Figure \ref{fig: votekit summary stats}).   The pairwise comparison graph shows  head-to-head margins between candidates.  For instance, 282 more voters ranked $W_2>C_1$ than $C_1>W_2$ in the preference profile used to make the figure.  Note that $W_1$, who has the most first-place votes by far, is also preferred head-to-head over all alternatives, making them the {\em Condorcet candidate} in this election.
+% For a given preference profile, basic \VK functions provide statistics and visuals for first-place votes, Borda count, and mention frequency, as well as head-to-head comparisons (see Figure \ref{fig: votekit summary stats}).   The pairwise comparison graph shows  head-to-head margins between candidates.  For instance, 282 more voters ranked $W_2>C_1$ than $C_1>W_2$ in the preference profile used to make the figure.  Note that $W_1$, who has the most first-place votes by far, is also preferred head-to-head over all alternatives, making them the {\em Condorcet candidate} in this election.
 
 \FloatBarrier
 %%%%%%%%%%%%%%%