Parsertime v2.0 - A Reimagined Experience

by Lucas Doell, Founder / Lead Developer

Welcome to the biggest update of Parsertime's lifetime! We're excited to announce Parsertime v2.0, a major overhaul of the platform with major new features and improvements to the overall experience. The first major update in about a year, Parsertime has been given a new coat of paint and a fresh new look for 2026. We’re excited to share some of the new features and see how they can take your competitive experience to the next level.

If you find any issues, please report them to us either through the use of the Report a Bug menu within the app or within our community Discord. Happy scrimming!

Team Statistics

Our most requested feature is finally here! Users are now able to view statistics for their team's performance and see detailed information on performance, such as strongest maps, heroes, rosters, and more.

A screenshot of the team statistics page.

Overview Tab — At-a-Glance Team Health

The Overview tab provides a snapshot of how your team is performing right now.

  • Quick Stats Card surfaces key indicators such as your last 10 games record, best day of the week, average fight duration, and first-pick success rate.
  • Team Roster Grid shows all active players with roles and core information for quick reference.
  • Recent Activity Calendar visualizes scrims and matches over time, helping you identify consistency, gaps, and burnout patterns.
  • Top Maps Card highlights your five most-played maps with win rates, revealing comfort picks.
  • Strengths & Weaknesses calls out your strongest map by win rate and your biggest blind spot (most played, lowest win rate).
  • Role Balance Radar compares Tank, Damage, and Support across eliminations, survivability, ultimate usage, and activity, producing a balance score and actionable role-level insights.

Performance Tab — Role and Composition Effectiveness

The Performance tab dives into how well each role and composition is actually working.

  • Role Performance Card breaks down win rates, per-10-minute stats, K/D, survivability, ultimate efficiency, and damage/healing/mitigation for Tank, Damage, and Support.
  • Best Role Trios Card identifies your most successful role distributions, ranking combinations by win rate and usage to show which team structures consistently perform best.

Heroes Tab — Hero Pool and Pick Strategy

The Heroes tab analyzes what your team plays and how effective those choices are.

  • Flexible Timeframe Selection lets you analyze recent form or long-term trends, with permissions controlling access to extended history.
  • Hero Pool Overview Card evaluates hero diversity, most-played heroes, win rates, role distribution, and overall pool depth.
  • Hero Pickrate Heatmap visualizes hero usage across maps and modes, revealing map-specific preferences and adaptation patterns.

Trends Tab — Performance Over Time

The Trends tab focuses on momentum and consistency.

  • Winrate Over Time Chart shows weekly and monthly performance trends with clear indicators of improvement or decline.
  • Recent Form Card summarizes short-term performance relative to your overall baseline.
  • Win/Loss Streaks Card tracks current and historical streaks, helping identify momentum swings and stability issues.

Maps Tab — Map and Mode Mastery

The Maps tab provides a complete breakdown of where your team thrives—or struggles.

  • Map Mode Performance Card analyzes win rates across Control, Escort, Hybrid, Push, Flashpoint, and more.
  • Map Winrate Gallery offers a visual overview of every map with playtime and performance indicators.
  • Player Map Performance Card highlights individual strengths by map, identifying specialists and informing lineup decisions.

Teamfights Tab — Fight Outcomes and Ultimate Economy

The Teamfights tab delivers deep insight into how fights are won and lost.

  • Team Fight Stats Card tracks fight win rate, first pick/death impact, first ultimate usage, dry fights, and average ultimates per fight.
  • Ultimate Economy Card evaluates coordination, efficiency, and timing across roles.
  • Win Probability Insights quantify how factors like first pick or ultimate advantage affect expected outcomes, turning fight data into actionable strategy.

With these new features, coaches and players alike have new tools to make data-driven decisions unlike anything previously available in the competitive Overwatch scene.

Composite Skill Rating (CSR)

One of the most significant additions is a composite skill rating (CSR) system. CSR is a skill rating derived from your statistical performance compared to the average player on a specific hero.

A screenshot of a player's hero skill rating.

For specific maps, each stat individually has its own skill rating as well. Your average per 10 minutes is compared to the average player, and the difference is converted into a skill rating. Each stat's skill rating is then combined into a composite skill rating that summarizes your overall performance on that hero.

For a player's profile, the CSR is calculated by their overall averages. All of their stats are averaged per 10 minutes and compared to the average player. Therefore, general outliers do not affect the CSR as heavily as they do for individual maps. Note that the profile SR is NOT an average of each map’s hero SR.

MVP Score

A player’s MVP score is calculated by their stats per 10 compared to the average player as well. This score is then converted to a range between ±100. Most scores will fall between ±50, with extreme outliers scoring above this range. This score is calculated differently than CSR, leading to a different distribution. MVP scores are recorded and can be viewed on your profile page.

The Formula

We calculate a Z-Score for each key statistic, which measures how many standard deviations you are above or below the average.

  • Stats are normalized to "per 10 minutes".
  • Positive stats (e.g., Eliminations) reward higher values.
  • Negative stats (e.g., Deaths, Damage Taken) reward lower values.

Role Weighting

Each role prioritizes different stats. For example:

  • Tank: Prioritizes low Deaths (30%), Eliminations (20%), and Solo Kills (15%).
  • Damage: Prioritizes Eliminations (30%), Final Blows (20%), and Damage Dealt (20%).
  • Support: Prioritizes Healing (35%) and low Deaths (25%). Specific heroes like Mercy have unique weightings.

Final CSR Calculation

The weighted Z-scores are summed and converted to an CSR scale centered at 2500 (average).

2500 + (Z_Score * (1250 / (1 + |Z_Score| / 3)))

This formula ensures that extreme outliers don't break the scale, while rewarding consistent high performance.

What is a good CSR?

In this leaderboard system, player scores (SR) are distributed along a bell curve, also known as a normal distribution. This means most players will have scores clustered around the average, and fewer players will have extremely high or low scores.

The average SR is 2500, representing the skill level of a typical player. Scores above 2500 are considered above average, while those below 2500 are below average.

The concept of the bell curve ensures that:

  • Most players will have SRs close to 2500.
  • A "good" SR is typically anything above the average (2500).
  • The further your SR is above 2500, the rarer and more impressive your ranking.

Standard deviation is used to measure how spread out the scores are around the average. If your SR is one standard deviation (about 300-400 SR) above 2500, you're already in roughly the top 16% of players. The higher your SR relative to 2500, the fewer players have achieved that score.

Leaderboard

A screenshot of the leaderboard page.
A leaderboard page has been added to display the top 50 players for each hero. This shows metrics like where the player is distributed on the bell curve of leaderboard scores, how their performance compares to others, and more.

How do I get on the leaderboard?

To appear on the leaderboard, you must meet the following criteria:

  • At least 10 maps played on the hero. This ensures the data is statistically significant and represents consistent performance.
  • At least 60 seconds of playtime per map. Brief hero swaps mid-game won't count toward your map total. Even if you don't appear in the top 50, your rank is still calculated and visible on your profile page, allowing you to track your progress over time.

Player Profiles

A major new feature is the addition of player profiles. Players can now add their Battletag to their account through the settings page, enabling the new profile page. This allows players to view their stats, receive titles for their achievements, and more.

A screenshot of a player's profile page.

The profile page now features new sections, such as:

  • Recent activity
  • Hero comparisons
  • Hero Mastery
  • Play Style charts
  • Personal Records
  • Achievements, with 16 new titles being awarded for merits such as hitting the highest rank on a hero leaderboard

To make sure you get the best experience, make sure to add your Battletag!

Technical Improvements

With the new update, we've made several technical improvements across the platform. The most significant change is that the parser now automatically fixes corrupted logs. This has been a pain point for many users and I'm happy to say that it has been resolved for a majority of issues. In particular, the parser now automatically fixes:

  • Corrupted stats that appear in the logs as ****
  • Gitched mercy_rez events that were caused by wave respawns If you still have issues, please reach out and we'll be happy to help you! You can visit our Discord for assistance or use the self-serve Debug Tool. That being said, we'd prefer that you reach out so that we can update the parser to fix the issue for the rest of the community.

There's also been a lot of work on the performance. In particular, we've reduced the loading times significantly across the platform. Users should notice a significant improvement when navigating to scrim pages, player pages, and more. We've also implemented pagination on the dashboard—users with a high quantity of scrims should see a huge improvement in loading times. There's also been upgrades to robustness and improved error logging across the board, which will help us identify and fix issues more quickly. Finally, we’ve also added a Terms of Service page for users who are curious about our policies.

Known Issues

  • Same issues as documented in v1.4 - please consult our known issues page within our documentation to stay updated.

Patch Notes

  • Overhauled UI for a new and improved user experience
    • A familiar yet new UI featuring many changes to the existing interface
  • Added a comprehensive Team Statistics page with deep, role-aware analytics across scrims and matches
    • Overview dashboard with quick stats, activity calendar, top maps, strengths & weaknesses, and role balance radar
    • Performance insights for Tank, Damage, and Support, including per-10 stats, win rates, survivability, and ultimate efficiency
    • Hero pool analysis with timeframe filtering, diversity scoring, and map-specific pickrate heatmaps
    • Trend tracking for win rates, recent form, and win/loss streaks over time
    • Detailed map and game-mode breakdowns, including player-by-map performance
    • Advanced teamfight and ultimate economy analysis with win-probability insights based on fight conditions
  • Added hero skill rating based on global averages and standard deviations, showing users their rank and percentile
  • Created an MVP score algorithm to highlight outliers in average performance
    • The MVP score is derived from a weighted average of specific metrics, such as eliminations, final blows, deaths, and more
    • Roughly weighs between -50 and 50, with higher values indicating a statistically significant difference from the global average
  • Created a new profile page to show off user stats, cosmetics such as titles and banners, and more
    • View a player's Fleta Deadlift stats (average over time, career high), average MVP score, first pick rate, and more
    • Premium subscribers can add a custom banner to their profile
    • There are 16 new titles achievable through actions such as being a beta tester, achieving the highest rank on a hero leaderboard, and more
    • Hero Mastery section now shows your hero skill rating for all heroes, not just top 3 most played heroes
    • Progression section shows a radar chart of your play style and a list of your career best stats
  • Added a new Leaderboard page to show the highest ranked players on each individual hero leaderboard, with charts for SR distribution and performance breakdowns
  • Basic plan and premium plan subscribers will now have a supporter heart next to their name on the profile page and in the profile hover card. Thank you for your support!
  • The parser now automatically fixes corrupted logs, reducing Invalid Log Format errors significantly
  • Added a notes section to the map page, allowing users to write Markdown notes for each map and record their thoughts
  • Improved navigation by prefetching scrims and map data on the dashboard, reducing user load times significantly
  • Added new colorblind features for improved accessibility, including custom colors
    • These new colorblind settings match the in-game Overwatch 2 colorblind settings
    • Custom color picker allows users to enter an exact hex code for team 1 and team 2 colors
  • Created an in-app notification system to inform users of changes made by their teammates
  • Overhauled the team page, showing user profiles instead of just the users' images and emails
  • Added an improved onboarding experience for new users, helping them get started with Parsertime quickly and easily
  • Implemented proper pagination support for the dashboard
    • Users with large amounts of scrims should see improved loading times
  • Added an audit log feature for admins to track changes made by users, improving customer support capabilities
  • Added various analytics charts for admins to see core metrics at a glance
  • Upgraded dependencies to the latest version for improved performance and security
  • Added a terms of service page for reference
  • Added support for Wuyang and Vendetta

Attribution

Special thanks to our beta testers jaypiz, MrPeppar, Myro, grantsco4117, ternary, Zydra, hampter0551, and Pudge for helping us make v2.0 an even better experience for everyone! If you're interested in becoming a beta tester, please join our Discord server and let us know!

Contributors

More articles

Releasing Parsertime's Dataset to the Community

Access an anonymized version of Parsertime's scrim dataset for data science and predictive modeling purposes.

Read more

Parsertime v1.4 - Going Global and the Future of Parsertime

We've updated Parsertime to version 1.4! This update includes internationalization support, compatibility with Hazard, and more.

Read more

Tell us about your idea

Points of contact