Data Driven Floor Leaders by Position: GDS Fantasy Aussie Rules 2025

Fantasy coaches spend too much time chasing ceilings. The ceiling is what wins you a single round. The floor is what stops you losing every other round. We used the full 2025 AFL season data to identify the most statistically reliable GDS Fantasy scorers at each position, based entirely on measurable numbers.

All data is sourced from the 2025 AFL season (Rounds 1 to GF). Only players with 50%+ time on ground per game and a minimum of 15 games are included in the primary analysis.

Defining "Floor" Mathematically

A single bad game is noise. The floor is the structural baseline a player produces when things are not going their way. We define it using five metrics:

1. 20th percentile score (P20): The score a player exceeds in 80% of their games. This is the primary floor measure.

2. Bottom 3 game average: The mean of a player's three lowest qualifying scores. Tests the true downside.

3. Standard deviation: Measures score volatility. Lower is tighter and more predictable. Higher means you never quite know what you're going to get.

4. Sub position threshold frequency: Percentage of games scoring below the positional median (DEF: 75, MID: 93, RUC: 96, FWD: 69 GDS points). Lower is better.

5. Median vs mean gap: A tight gap indicates consistent output without extreme scores inflating or deflating the average.

These five metrics together isolate players whose worst games are still serviceable and whose scoring distribution is narrow and predictable.

Defenders: Best GDS Fantasy Floor Profile

Lachie Whitfield

Whitfield posted the strongest defensive floor profile in the 2025 dataset. His 20th percentile score of 111.6 GDS points is the second highest among all defenders, but his combination of low volatility and elite downside protection makes him the standout. An important note is we have excluded the injury affected game from this report. An injury is something that can't be predicted and therefore shouldnt be built into modelling.

In qualifying games, Whitfield never scored below the positional median of 75 GDS points. His standard deviation of 20.2 is the lowest among the top four defenders, meaning his scores cluster more tightly around his average. His bottom 3 game average of 98.7 is remarkable; even in his worst three outings, he still averaged a score that would rank as elite for most defenders.

Consistency analysis: Whitfield's median (133.5) actually exceeds his mean (130.9), a gap of just 2.6 points. This indicates his distribution is slightly left skewed with minimal downside drag. Approximately 80% of his games fell between 112 and 151 GDS points (mean plus or minus one standard deviation). His last 5 game median of 143 shows no sign of deterioration.

Why not Wanganeen-Milera? NWM posted a higher P20 (113.4) and average (135.4), but his standard deviation is 33% higher (27.0 vs 20.2) and his bottom 3 average is 4 points lower. His floor is excellent, but Whitfield's is more compressed and predictable.

Rory Laird deserves a mention for the lowest standard deviation among qualifying defenders at 17.0, though his overall scoring level (avg 107.8) is lower. His median to mean gap is just 4.2 points, suggesting highly symmetric output. His P20 of 92.2 is still strong for a set and forget defensive pick.

Platform note: Whitfield averaged 109.7 AFL Fantasy and 116.0 SuperCoach points. His floor strength translates across all three platforms, making him a true multi platform consistency pick.

Risk factors: Whitfield is 31 (turns 32 in 2026). Age related decline is a monitoring point. No measurable weather sensitivity was detected so nothing to worry about there. His last 5 median of 143 suggests current form is, if anything, above baseline.

Midfielders: Best GDS Fantasy Floor Profile

Bailey Smith

Smith's 2025 season produced the highest P20 score of any midfielder in the dataset at 120 GDS points. That alone would be noteworthy. Combined with a bottom 3 average of 96.3 and zero games below 70, the floor profile is the strongest in the competition.

Smith dipped below the positional median of 93 GDS points in just 1 of 23 games (4.3%). His worst score all season was 85. For context, 85 GDS points would be above the positional median for defenders and forwards. His floor effectively operates at the ceiling of other positions.

Consistency analysis: Smith's median (144) exceeds his mean (137.4) by 6.6 points. This is a meaningful gap that suggests occasional lower scores pull the average down, but the typical game output sits higher. Roughly 80% of his games fell between 114 and 161 GDS points. His last 5 median of 127 is solid but slightly below his season median, something to monitor heading into 2026.

Jordan Dawson is the runner up and arguably the most complete floor profile when factoring sample size (25 games). His P20 of 106.4 ranks second, his bottom 3 average of 96 nearly matches Smith's, and his sub position threshold rate of just 4% (1 of 25 games) is the best among all midfielders. His lowest single score was 90, the highest minimum of any midfielder with 20+ games.

Finn Callaghan rounds out the top tier with a P20 of 105.0 and the second lowest standard deviation (20.8) among the top group. His last 5 median of 142 suggests he finished the season in strong form.

Platform note: Smith averaged 115.0 AFL Fantasy and 116.4 SuperCoach. Dawson averaged 108.9 AFL Fantasy and 110.4 SuperCoach. Both are elite across all platforms. Callaghan's SuperCoach average (111.6) outperformed his GDS average, making him particularly strong in that format.

Risk factors: Smith's last 5 median (127) is 17 points below his season median (144). This could indicate late season regression or matchup variation. Monitor early 2026 data before locking in as a set and forget. Dawson shows no such trend; his output was consistent throughout.

Rucks: Best GDS Fantasy Floor Profile

Tim English

English leads the ruck division with a P20 of 109.4 GDS points and the best sub threshold rate among premium rucks. He scored below the positional median (96 GDS) in just 1 of 23 games (4.3%), a rate that matches or betters most elite midfielders.

English's floor advantage over Max Gawn is notable. While Gawn's average (127.4) and median (130) are higher, English's P20 is 5.8 points higher and his bottom 3 average is 3.3 points higher. More importantly, English's lowest single score (77) is 8 points above Gawn's (69), and his sub threshold rate is half of Gawn's (4.3% vs 8.7%).

Consistency analysis: English's mean (124.4) exceeds his median (118) by 6.4 points, indicating occasional high ceiling games push the average up, but the typical output sits at 118. This is still a strong baseline. Roughly 80% of his games fell between 102 and 147 GDS points. His last 5 median of 138 indicates a strong finish to the season.

Darcy Cameron stands out as the safest downside ruck. His lowest score (81) is the highest minimum among all qualifying rucks, and he never scored below 70. However, his P20 of 95.6 and sub threshold rate of 20.0% are significantly worse than English's, meaning his typical bad games are worse even though his absolute worst game is better protected.

Platform note: English averaged 110.7 AFL Fantasy and 111.0 SuperCoach. Gawn's SuperCoach average (127.4) is substantially higher due to the platform's hitout weighting. If you play SuperCoach, Gawn's floor profile is stronger on that platform specifically.

Risk factors: Ruck scoring is inherently more volatile than other positions due to game script and opponent ruck quality. English's standard deviation of 22.2 is the lowest among the top four rucks. No measurable weather sensitivity detected. His last 5 median of 138 shows strong current form.

Forwards: Best GDS Fantasy Floor Profile

Harry Sheezel

Sheezel's P20 of 109.8 GDS points is not just the best among forwards. It would rank as the best floor score among defenders and would sit inside the top five among midfielders. His floor profile is genuinely position breaking.

Sheezel didn't scored below the forward positional median (69 GDS) across 23 games which should surprise nobody. His worst single game was 75, which is still above the positional median. He is the only forward in the dataset whose worst individual score exceeds the positional median.

Consistency analysis: Sheezel's median (125) sits just below his mean (126.0), a gap of 1.0 point. This is the tightest median to mean gap among all top forward candidates, indicating a near perfectly symmetric scoring distribution. Approximately 80% of his games fell between 99 and 153 GDS points. His last 5 median of 129 confirms sustained output.

Christian Petracca is the runner up and makes a compelling case as the most predictable forward in the dataset. His standard deviation of 15.9 is by far the lowest among qualifying forwards (next best is Mannagh at 23.3). His P20 of 91.0 is lower than Sheezel's, but his scoring range is far tighter. Petracca scored between 79 and 134 in every single game. He also never dipped below the positional median. The trade off is his average (103.9) is 22 points lower than Sheezel's. If you want pure predictability over scoring level, Petracca is the pick.

Izak Rankine warrants a note: 22 games, P20 of 84.2, zero games below the positional median, lowest score 78. His sub threshold rate of 0% matches Sheezel and Petracca, making him the third forward who never dipped below median. Standard deviation of 19.8 is tidy for a forward.

Platform note: Sheezel averaged 109.2 AFL Fantasy and 107.3 SuperCoach. His GDS average (126.0) is significantly higher, suggesting the GDS scoring system is particularly generous to his stat profile. Petracca's SuperCoach average (96.5) was relatively higher than his GDS (103.9), so the gap closes on that platform.

Risk factors: Sheezel's standard deviation of 26.8 is higher than Petracca's (15.9), meaning his week to week range is wider despite the higher baseline. His bottom 3 average of 90.3 is 9 points above Petracca's (81.3), so even in his bad games the output is strong. No trend deterioration detected in last 5 data.

Cross Position Comparison

The strongest absolute floor belongs to Bailey Smith (MID) with a P20 of 120.0 GDS points. However, the cleanest floor profile when combining all metrics belongs to Lachie Whitfield: the lowest standard deviation (20.2), the highest bottom 3 average (98.7), and zero games below the positional median across 22 games.

By position, midfielders produce the highest absolute floors but also the highest variance among top candidates. Defenders and forwards produced the best sub threshold rates, with both Whitfield and Sheezel posting 0% below their respective positional medians.

Rucks are the most volatile position for floor analysis. Even the best ruck (English, std dev 22.2) produces wider score distributions than the best defender (Whitfield, 20.2) or the most predictable forward (Petracca, 15.9). Ruck coaches should expect more week to week variance even from premium selections.

Practical Application for GDS Fantasy

Floor analysis is most valuable in two GDS Fantasy scenarios. First, when selecting your weekly squad, prioritising high floor players reduces the risk of a catastrophic low score dragging down your team total. Second, when evaluating purchases, a player with a higher ceiling but lower floor introduces more variance into your lineup, which may not be desirable if you are protecting a ranking position.

The five players identified here (Smith, Whitfield, English, Sheezel, and Petracca as a value alternative) represent the statistical backbone of a volatility minimised GDS Fantasy squad in 2025. Whether those floor profiles hold into 2026 depends on role continuity, fitness, and team context. Monitor early season data before committing to season long structures.

Methodology Note

All analysis uses 2025 AFL season data (Rounds 1 to GF). Minimum sample size for primary analysis: 15 games with 50%+ time on ground. Positions based on 2026 GDS Fantasy mappings. Positional medians calculated across all qualifying games at each position (DEF: 75, MID: 93, RUC: 96, FWD: 69 GDS points). P20 calculated using linear interpolation.